System and method for operating autonomous vehicle

Abstract

An autonomous vehicle (AV) includes features that allow the AV to comply with applicable regulations and laws for performing safe driving operations. An example system for an AV includes acquiring, by a computer located in the AV, an image from a camera located on the AV, wherein the image characterizes an area towards which the AV is driven on a road or a lane on a highway; determining that the pedestrian or the rider is located beside the road or the lane on the road according to the image; and in response to the determination, performing a driving operation on the AV, such as turning from a center of the lane to a first side of the lane away from the center of the lane and away from the position of the pedestrian or rider, and / or decelerating the AV in response to certain conditions.

Description

[0001] Priority claims and related patent applications

[0002] This patent document claims U.S. Provisional Application No. 63 / 181,779, filed on April 29, 2021, entitled “System and Method For An Autonomous Vehicle,” U.S. Provisional Application No. 63 / 181,782, filed on April 29, 2021, entitled “System and Method For An Autonomous Vehicle,” U.S. Provisional Application No. 63 / 181,786, filed on April 29, 2021, entitled “System and Method For An Autonomous Vehicle,” U.S. Provisional Application No. 63 / 216,357, filed on June 29, 2021, entitled “System and Method For An Autonomous Vehicle,” The present invention claims priority to and the benefits of U.S. Provisional Application No. 63 / 216,358, filed on September 29, 2021, entitled “System and Method For An Autonomous Vehicle”, U.S. Provisional Application No. 63 / 250,212, filed on September 29, 2021, entitled “System and Method For An Autonomous Vehicle”, and U.S. Provisional Application No. 63 / 255,839, filed on October 14, 2021, entitled “System and Method For An Autonomous Vehicle”. The entire disclosures of the foregoing applications are incorporated herein by reference as a part of the disclosure of this application. Technical Field

[0003] The present disclosure relates generally to autonomous vehicles. More particularly, the present disclosure relates to properly operating an autonomous vehicle (AV) on public roads, highways, and locations where there are other vehicles or pedestrians. Background Art

[0004] Autonomous vehicle technology can provide a vehicle that safely navigates to a destination with limited or no driver assistance. Safe navigation of an autonomous vehicle (AV) from one point to another may include the ability to signal other vehicles, navigate around other vehicles in the shoulder or emergency lane, change lanes, properly offset in lanes, and navigate all sections or types of highway lanes. Autonomous vehicle technology can enable an AV to operate without requiring extensive learning or training from surrounding drivers by ensuring that the AV can operate safely in a manner that is obvious, logical, or familiar to surrounding drivers and pedestrians. Summary of the invention

[0005] Systems and methods are described herein that can allow an autonomous vehicle (AV) to navigate from a first point to a second point. In some embodiments, the AV can navigate from the first point to the second point without a human driver in the AV and comply with instructions for safe and legal operations.

[0006] A first example method of operating an autonomous vehicle includes: acquiring, by a computer located in the autonomous vehicle, an image from a camera located on the autonomous vehicle, wherein the image represents an area toward which the autonomous vehicle is driving on a lane of a road or highway; determining, based on the image, that a pedestrian or cyclist is located next to the lane on the road or highway; in response to the determination, operating the autonomous vehicle to turn from the center of the lane to a first side of the lane, the first side of the lane being away from the center of the lane and away from the location of the pedestrian or cyclist; and in response to the determination, operating the autonomous vehicle to reduce the speed of the autonomous vehicle to below a first threshold speed value in response to determining that a lateral distance from the autonomous vehicle to the pedestrian or cyclist is within a first distance set and that a current speed of the autonomous vehicle is greater than a first threshold speed value.

[0007] In some embodiments, the autonomous vehicle is caused to reduce the speed of the autonomous vehicle by comparing the lateral distance from the autonomous vehicle to the pedestrian or cyclist and the current speed of the autonomous vehicle to a table including a plurality of distance sets and a plurality of threshold speed values, wherein the plurality of distance sets include a first distance set and a second distance set greater than or equal to the first distance set, wherein the plurality of threshold speed values ​​include a first threshold speed value and a second threshold speed value greater than the first threshold, and wherein the first distance set and the second distance set correspond to the first threshold speed value and the second threshold speed value, respectively. In some embodiments, the first threshold speed value is a minimum of a first predetermined speed value and a first speed value, the first speed value being obtained by subtracting a specific speed from a speed limit, and the second threshold speed value is a minimum of the second predetermined speed value and the first speed value.

[0008] In some embodiments, the method further includes: in response to determining that the lateral distance from the autonomous vehicle to the pedestrian or cyclist is greater than a third distance set, operating the autonomous vehicle to maintain the speed of the autonomous vehicle, the third distance set being greater than or equal to the second distance set. In some embodiments, the method further includes: in response to determining the presence of an emergency vehicle on the road or highway based on the image: operating the autonomous vehicle to reduce the speed of the autonomous vehicle to below a third threshold speed value in response to determining that the lateral distance from the autonomous vehicle to the pedestrian or cyclist is within the first distance set and the current speed of the autonomous vehicle is greater than a third threshold speed value, wherein the third threshold speed value is the minimum of the first threshold speed value and the maximum overtaking speed value.

[0009] In some embodiments, the maximum overtaking speed value is a specific speed that is less than a speed value, wherein the speed value is based at least on a speed limit of a road or highway and whether the autonomous vehicle is operating on the road or highway. In some embodiments, the method further includes: operating the autonomous vehicle to overtake a pedestrian or cyclist by maintaining a minimum lateral distance between the autonomous vehicle and the pedestrian or cyclist, wherein the minimum lateral distance is a predetermined distance from a side of the autonomous vehicle farthest from the pedestrian or cyclist to the position of the pedestrian or cyclist. In some embodiments, the pedestrian or cyclist is determined based on an image when a first distance from a first position of the autonomous vehicle to a second position of the pedestrian or cyclist is greater than or equal to a stopping distance of the autonomous vehicle, and wherein the stopping distance is a second distance required for the autonomous vehicle to come to a complete stop.

[0010] A second example method of operating an autonomous vehicle includes determining, by a computer located in the autonomous vehicle, that the autonomous vehicle is decelerating when the autonomous vehicle is at a first position on a road, the first position being within a predetermined distance of a second position at which the autonomous vehicle is to perform a steering maneuver; and in response to the determination and in response to determining that the turn signal is not on, operating the turn signal to turn on at the first time.

[0011] In some embodiments, the method further includes sending an instruction in response to determining that the turn signal is not working or operating, the instruction causing the autonomous vehicle to turn to the side of the road along the trajectory and apply brakes. In some embodiments, the turn signal is caused to be turned on for a first time length at a first time, and wherein the method further includes: performing a first determination to determine that the first time length overlaps with a second time length, the second time length being associated with a second steering maneuver after the steering maneuver; performing a second determination to determine that the second steering maneuver is in the same direction as the steering maneuver; and in response to the first determination and the second determination, operating the turn signal to remain on during the first time length and the second time length. In some embodiments, the predetermined distance is based on laws or regulations of the region or state in which the autonomous vehicle is operating.

[0012] In some embodiments, the method also includes: performing a determination to determine that a second location at which the autonomous vehicle is to perform a steering maneuver is adjacent to, within a certain distance of, or passing through the intersection; and wherein in response to the determination, in response to determining that the turn signal is not turned on, and in response to determining that the rear of the autonomous vehicle passes through the middle of the intersection, causing the turn signal to be turned on.

[0013] A third example method for operating an autonomous vehicle includes determining, by a computer located in the autonomous vehicle, that a lane of a road on which the autonomous vehicle is operating includes a curved portion, the curved portion having a minimum radius greater than or equal to a predetermined threshold; and when the autonomous vehicle is driving through the curved portion of the road, operating the autonomous vehicle traveling on the curved portion to move toward one side of the lane and away from a center of the lane, wherein the side of the lane is a side that curves outward, and wherein the autonomous vehicle is caused to move toward one side of the lane until a predetermined threshold distance from the center of the lane.

[0014] In some embodiments, the method further includes: in response to determining that an additional curved portion of the road has a superelevation greater than a predetermined threshold amount, operating the autonomous vehicle to avoid driving on the additional curved portion, wherein the superelevation describes an upward angle formed by the additional curved portion tilting upward relative to the flat surface. In some embodiments, the predetermined threshold amount is a value between 5% and 15%. In some embodiments, the method further includes: operating the autonomous vehicle to accelerate or decelerate less than or equal to a predetermined rate in the curved portion of the road. In some embodiments, the method further includes: operating the autonomous vehicle to reduce the speed of the autonomous vehicle in response to determining that the time to collision (TTC) value of the autonomous vehicle when operating in the curved portion is greater than a predetermined amount of time, wherein the TTC value indicates the amount of time of visibility provided by one or more cameras on the autonomous vehicle.

[0015] A fourth example method for operating an autonomous vehicle includes determining, by a computer located in the autonomous vehicle, that an emergency vehicle is within a predetermined distance of a first position of the autonomous vehicle operating in a lane on a road; and in response to the determination, operating the autonomous vehicle to turn from a center of the lane toward a first side of the lane, the first side of the lane being away from the center of the lane and away from a second position of the emergency vehicle, wherein the autonomous vehicle is caused to turn toward the first side until a lateral distance between the emergency vehicle and the autonomous vehicle is greater than or equal to the predetermined distance.

[0016] In some embodiments, the method further comprises: in response to determining that the line separating the lane and the second lane adjacent to the lane comprises a white dashed line, a yellow dashed line, or a white solid line, causing the autonomous vehicle to turn toward the first side of the lane and turn into the second lane. In some embodiments, the method further comprises: in response to determining that the emergency vehicle is within a predetermined distance of the first position of the autonomous vehicle, and in response to determining that a lane change operation of the autonomous vehicle is not possible, sending an instruction that causes the autonomous vehicle to apply braking, or to decelerate the autonomous vehicle to a speed less than a threshold speed value.

[0017] In some embodiments, the threshold speed value is based on the rules of the region or state or region in which the autonomous vehicle is located. In some embodiments, the threshold speed value is based on the speed limit of the first location where the autonomous vehicle is operating and the lateral distance between the emergency vehicle and the autonomous vehicle. In some embodiments, the threshold is based on the speed limit of the first location where the autonomous vehicle is operating and the lateral distance between the emergency vehicle and the autonomous vehicle. In some embodiments, the autonomous vehicle operates to turn from the center of the lane toward the first side of the lane, and causes the autonomous vehicle to apply braking, or decelerate the autonomous vehicle to a speed less than the threshold speed value in response to the following circumstances: determining that the emergency vehicle is operating in another lane adjacent to the lane on which the autonomous vehicle is operating; and determining that a lane change operation of the autonomous vehicle is impossible. In some embodiments, the method also includes: operating the autonomous vehicle to accelerate only for lane change or for performing an avoidance maneuver in response to determining that the emergency vehicle is approaching the autonomous vehicle and the emergency vehicle is maneuvering in another lane adjacent to the lane on which the autonomous vehicle is operating.

[0018] In some embodiments, the system further comprises a sensor subsystem, the sensor subsystem comprising a camera, a temperature sensor, an inertial sensor (IMU), a global positioning system, a light sensor, a LIDAR system, a radar system, and wireless communication, and wherein the computer located in the autonomous vehicle is configured to utilize data from any sensor subsystem to perform determinations and operations. In some embodiments, the system further comprises a vehicle control subsystem in operative communication with the computer located in the autonomous vehicle, wherein the processor is configured to communicate with the vehicle control subsystem to perform a method that causes the autonomous vehicle to turn from the center of the lane toward a first side of the lane, and causes the autonomous vehicle to apply braking, or decelerate the autonomous vehicle to a speed less than a threshold speed value, in response to: determining that an emergency vehicle is operating in another lane that is immediately adjacent to the lane on which the autonomous vehicle is operating; and determining that a lane change operation of the autonomous vehicle is not possible. In some embodiments, the system also includes a vehicle control subsystem operably connected to a computer located in the autonomous vehicle, wherein the processor is configured to execute the method, the method also including: operating the autonomous vehicle via the vehicle control system to accelerate solely for the purpose of changing lanes or to perform an evasive maneuver in response to determining that an emergency vehicle is approaching the autonomous vehicle and that the emergency vehicle is operating in another lane that is immediately adjacent to the lane in which the autonomous vehicle is operating.

[0019] In some embodiments, the threshold speed value is based on: the rules of the region, state, or area in which the autonomous vehicle is located; and the speed limit of the first location where the autonomous vehicle is operating and the lateral distance between the emergency vehicle and the autonomous vehicle. In some embodiments, the method also includes: for the emergency vehicle transitioning to the emergency lane vehicle, the autonomous vehicle changing lanes to move away from the lane adjacent to the emergency lane; and slowing down and matching the speed of the emergency vehicle transitioning to the emergency lane by the autonomous vehicle until the emergency vehicle exits the current lane of travel of the autonomous vehicle. In some embodiments, the autonomous vehicle uses onboard sensors to detect any of the following to identify the emergency vehicle as a vehicle transitioning to the emergency lane: the use of a turn signal by the emergency vehicle to indicate a direction toward the shoulder of the road; a deviation or change in trajectory of the emergency vehicle; the activation of flashing lights indicating an emergency vehicle, rescue vehicle, or law enforcement vehicle; a change in speed of the emergency vehicle; and a direct communication from the emergency vehicle to the autonomous vehicle indicating the emergency vehicle's intent to move to the emergency lane or shoulder of the road.

[0020] A fifth example method for operating an autonomous vehicle includes: acquiring, by a computer located in the autonomous vehicle, an image from a camera located on the autonomous vehicle, wherein the image represents an area on a road toward which the autonomous vehicle is driving; performing a first determination based on the image to determine that the vehicle is stopped in the area for reasons unrelated to traffic congestion, traffic signals, or traffic signs; performing a second determination to determine that the autonomous vehicle is expected to be driven within a predetermined lateral distance from the vehicle; and, in response to the first determination and the second determination, operating the autonomous vehicle to automatically operate the vehicle at a speed that is less than a maximum speed allowed for the autonomous vehicle to overtake or exceed the vehicle.

[0021] In some embodiments, the method further includes: operating the autonomous vehicle to turn from the first lane to a second lane adjacent to the first lane, the distance from the vehicle being greater than or equal to the predetermined lateral distance from the vehicle. In some embodiments, the method further includes: performing a third determination to determine that the vehicle is stopped in the same lane as the autonomous vehicle; performing a fourth determination to determine that the autonomous vehicle cannot change lanes; and in response to the third determination and the fourth determination, operating the autonomous vehicle to apply braking to stop the autonomous vehicle. In some embodiments, the maximum speed is based on whether the autonomous vehicle is operating on a local road or highway.

[0022] A sixth example method for operating an autonomous vehicle includes: acquiring, by a computer located in the autonomous vehicle, an image from a camera located on the autonomous vehicle, wherein the image represents an area toward which the autonomous vehicle is driving on an entrance ramp of a highway; determining, based on the image, that the area includes a merging section on a lane on the highway, and the autonomous vehicle is expected to merge onto the highway in the merging section; operating a turn signal to turn on in response to the determination, wherein the turn signal indicates that the autonomous vehicle is expected to merge into a lane on the highway from the entrance ramp; and operating the autonomous vehicle to turn from the entrance ramp of the highway to the merging section on the lane of the highway in response to the determination and in response to the turn signal being turned on.

[0023] In some embodiments, the total length of the merging segment includes the length of the autonomous driving segment, the first minimum distance allowed between the autonomous driving vehicle and a first vehicle expected to be located in front of the autonomous driving vehicle, and the second minimum distance allowed between the autonomous driving vehicle and a second vehicle expected to be located behind the autonomous driving vehicle. In some embodiments, the method further includes: performing a first determination to determine that the length of the merging segment is decreasing; and in response to the first determination, operating the autonomous driving vehicle to apply brakes to stop the autonomous driving vehicle. In some embodiments, the method further includes: in response to the determination, performing a second determination of the trajectory followed by the autonomous driving vehicle from the entrance ramp to the merging segment, wherein the trajectory avoids the autonomous driving vehicle from entering the lane splitting angle area. In some embodiments, the image is acquired by the autonomous driving vehicle when it is determined that there is no additional merging segment in the prior image of the other area toward which the autonomous driving vehicle is driving from the camera and when the autonomous driving vehicle is operated to move forward slowly on the highway, wherein the prior image is acquired in time before the time when the image is acquired from the camera. In some embodiments, the autonomous driving vehicle moves forward slowly at a speed less than a predetermined speed.

[0024] A seventh example method for operating an autonomous vehicle includes: acquiring, by a computer located in the autonomous vehicle, a set of images from a first camera located on the autonomous vehicle, wherein the set of images represents an area adjacent to a lane on a road on which the autonomous vehicle is driving; acquiring, by the computer, an image from a second camera located on the autonomous vehicle, wherein the image representation includes an additional area of ​​the lane on which the autonomous vehicle is driving; performing a first determination based on the set of images to determine that the vehicle is driving adjacent to the autonomous vehicle for a length of time; performing a second determination based on the image or the set of images to determine a level of risk associated with an autonomous vehicle driving parallel to the vehicle; performing a third determination in response to the first determination and the second determination to determine that the length of time is greater than a predetermined time period; and operating the autonomous vehicle to accelerate or decelerate in response to the third determination.

[0025] In some embodiments, performing the second determination includes: determining that the risk level is low in response to determining from the image that the width of the lane is within a range of predefined standard widths for a standard lane, and in response to determining that a trajectory for the autonomous vehicle to turn away from the center of the lane to one side of the lane is available, wherein the predetermined time period is associated with the low risk level. In some embodiments, performing the second determination includes: determining that the risk level is medium in response to determining from the image that the width of the lane is less than a range of predefined standard widths for a standard lane, or determining that a trajectory for the autonomous vehicle to turn away from the center of the lane to one side of the lane is unavailable, or determining that the lane includes a curved portion; wherein the predetermined time period is associated with the medium risk level.

[0026] In some embodiments, performing the second determination includes: determining that the risk level is high in response to determining that the autonomous vehicle is parallel to the turning vehicle or within a certain distance parallel to the turning vehicle based on the image set; and wherein the method further includes: operating the autonomous vehicle to accelerate or decelerate or change lanes in response to the determination. In some embodiments, the method further includes: determining that the length of the vehicle is greater than a predetermined length based on at least one image in the image set; and in response to the determination, operating the autonomous vehicle to turn away from the center of the lane to one side of the lane.

[0027] In yet another exemplary aspect, a system for operating an autonomous vehicle includes a computer including a processor configured to execute the above method and the method described in this patent document.

[0028] In another exemplary aspect, the above method and the method described in this patent document are embodied in a non-transitory computer-readable storage medium. The non-transitory computer-readable storage medium includes code that, when executed by a processor, causes the processor to perform the method described in this patent document.

[0029] In another exemplary embodiment, a device configured or operable to perform the above method is disclosed. In yet another exemplary embodiment, a system is disclosed that includes a computer located in a vehicle, the computer including a processor configured to implement the above method.

[0030] The above-described and other aspects and embodiments thereof are described in more detail in the drawings, the description and the claims. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] For a more complete understanding of the present disclosure, reference is now made to the following brief description of the drawings, taken in conjunction with the accompanying drawings and detailed description, wherein like reference numerals represent like parts.

[0032] Figure 1 A block diagram of an example vehicle ecosystem for an autonomous vehicle is shown.

[0033] Figure 2 A flow chart is shown for safely operating an autonomous vehicle taking into account the health of the autonomous vehicle and / or the surrounding environment.

[0034] Figure 3 A system including one or more autonomous vehicles, a control center or supervisory system with an operator (e.g., a remote center operator (RCO)), and an interface for third-party interaction is shown.

[0035] FIG. 4A to FIG. 4CThree example scenarios are shown of an emergency vehicle approaching an autonomous vehicle on a road.

[0036] FIG. 5A to FIG. 5C An example scenario is shown in which an emergency vehicle approaches from the left side of an autonomous vehicle on a road.

[0037] FIG. 5D to FIG. 5E An example scenario is shown of an emergency vehicle approaching an autonomous vehicle, where neither the emergency vehicle nor the autonomous vehicle is in the rightmost lane of the road.

[0038] Figure 6 Shown is a suggested following distance between an autonomous vehicle and a non-player character (NPC) vehicle.

[0039] Figure 7 An example scenario is shown in which a vehicle goes off track during a 90 degree turn.

[0040] Figure 8 An example scenario is shown of an autonomous vehicle returning to the center of a lane after performing a lane deviation maneuver when one or more vehicles are located in another lane adjacent to the lane in which the autonomous vehicle is operating.

[0041] Fig. 9 An example merging area for a k-ramp is shown.

[0042] Fig.10 An example scenario is shown in which an automated vehicle may avoid a cyclist when approaching a right-turn lane or a reduced lane.

[0043] Fig.11 An identification of the gestures determined by the autonomous vehicle and the corresponding meaning is shown, so that the autonomous vehicle can react to the cyclist's gesture signals.

[0044] Fig.12 An example of a wide lane merge area is shown.

[0045] Fig.13 An example scenario of driving maneuvers performed by an autonomous vehicle traveling alongside a disabled or incapacitated vehicle is shown.

[0046] Fig.14 An example acceleration stop zone is shown that may be adjacent to the location of a disabled or incapacitated vehicle.

[0047] Fig.15 Example driving-related maneuvers performed by an automated driving vehicle operating on a multi-lane entrance ramp on a highway are shown.

[0048] Fig.16An example flow chart of autonomous driving operations performed by a vehicle operating on a road or highway that includes pedestrians and / or cyclists is shown.

[0049] Fig.17 An example flow chart of an autonomous driving operation performed by a vehicle to operate a turn signal is shown.

[0050] Fig.18 An example flow chart of autonomous driving operations performed by a vehicle to operate on a curved section of a road is shown.

[0051] Fig.19 An example flow chart of autonomous driving operations performed by a vehicle to operate on a road with emergency vehicles is shown.

[0052] Fig. 20 An example flow chart of autonomous driving operations performed by a vehicle to operate on a road with stopped vehicles is shown.

[0053] Fig.21 An example flow chart of an autonomous driving maneuver performed by a vehicle to merge onto a highway is shown.

[0054] Fig. 22 An example flow chart of an autonomous driving operation performed by a vehicle to operate adjacent to another vehicle is shown. DETAILED DESCRIPTION

[0055] Vehicles traversing highways and roads are legally required to comply with regulations and ordinances in the course of safe operation of the vehicle. For autonomous vehicles (AVs), and in particular for autonomous semi-trailer trucks, the ability to identify faults in their systems and safely stop can allow for legal and safe operation of the vehicle. Systems and methods for safe and legal operation of an autonomous vehicle on a roadway, including the execution of maneuvers that enable the autonomous vehicle to comply with laws while signaling surrounding vehicles of their condition, are described in detail below.

[0056] This patent document describes an example vehicle ecosystem for an autonomous vehicle and driving-related operations of an autonomous vehicle in Section I below. Section II describes a control center or supervision system for one or more autonomous vehicles. Sections III to XXXI describe operations performed by autonomous vehicles in various scenarios. The example titles of the following sections are used to facilitate understanding of the disclosed subject matter and do not limit the scope of the claimed subject matter in any way. Accordingly, one or more features of one example section may be combined with one or more features of another example section.

[0057] This patent document uses many abbreviations and uncommon terms. For example, "GNSS" or "GPS" may refer to a satellite navigation system; when referring to emergency vehicles such as police cars, ambulances, fire trucks, tow trucks, etc., the abbreviation "EV" may be used; the acronym "TTC" indicates "time to collision"; "NPC" refers to a non-player character and may include non-player characters. Figure 1 Any vehicle other than the autonomous vehicle in the . For example, any surrounding vehicles, motorcycles, bicycles, etc. that are manually driven or autonomously driven and may not communicate with the autonomous vehicle can be considered as NPCs; "k-ramp" means such as Fig. 9 A particular configuration of highway on / off ramps is shown; "STV" indicates a stopped vehicle; and "ELV" may indicate a disabled vehicle or an incapacitated vehicle, such as a disabled vehicle at the side of a road.

[0058] I. Example Ecosystem for Autonomous Vehicles

[0059] Figure 1 A system 100 including an autonomous vehicle 105 is shown. The autonomous vehicle 105 may include a tractor for a semi-trailer truck. The autonomous vehicle 105 includes a plurality of vehicle subsystems 140 and an onboard control computer 150. The plurality of vehicle subsystems 140 include a vehicle drive subsystem 142, a vehicle sensor subsystem 144, and a vehicle control subsystem. An engine or motor, wheels and tires, a transmission, an electrical subsystem, and a power subsystem may be included in the vehicle drive subsystem. The engine of the autonomous truck may be an internal combustion engine, a fuel cell powered electric engine, a battery powered electric engine, a hybrid engine, or any other type of engine capable of moving the wheels on which the autonomous vehicle 105 moves. The autonomous vehicle 105 has a plurality of motors or actuators to drive the wheels of the vehicle, so that the vehicle drive subsystem 142 includes two or more electric drive motors. The transmission may include a continuously variable transmission or a set number of gears that convert the power generated by the engine into the force to drive the wheels. The vehicle drive subsystem may include an electrical system that monitors and controls the current distribution of components within the system, including pumps, fans, and actuators. The power subsystem of a vehicle drive subsystem may include components that regulate the vehicle's power source.

[0060] The vehicle sensor subsystem 144 may include sensors for general operation of the autonomous vehicle 105, including those sensors that would indicate a malfunction in the autonomous vehicle or another reason for the autonomous vehicle to perform a limited or minimal risk condition (MRC) maneuver. Figure 1The MRC maneuver may be performed by sending instructions that cause the autonomous vehicle to turn to one side of the road along the trajectory and apply brakes so that the autonomous vehicle can safely stop on the side of the road. Sensors for general operation of autonomous vehicles may include cameras, temperature sensors, inertial sensors (IMU), global positioning systems, light sensors, LIDAR systems, radar systems, and wireless communications.

[0061] A sound detection array, such as a microphone or microphone array, may be included in the vehicle sensor subsystem 144. The microphones of the sound detection array are configured to receive audio indications of the presence of an official or instructions from an official, including sirens and commands such as "pull over". These microphones are mounted or located on the exterior of the vehicle, specifically, outside the towing vehicle portion of the autonomous vehicle 105. The microphones used may be of any suitable type, mounted so that they are effective both when the autonomous vehicle 105 is stationary and when the autonomous vehicle 105 is moving at normal driving speeds.

[0062] The cameras included in the vehicle sensor subsystem 144 may be rear-facing so that flashing lights from emergency vehicles may be observed from all around the self-driving truck 105. These cameras may include video cameras, cameras with filters for specific wavelengths, and any other cameras suitable for detecting emergency vehicle lights based on color, flash, or both color and flash.

[0063] The vehicle control subsystem 146 may be configured to control the operation of the autonomous vehicle or truck 105 and its components. Accordingly, the vehicle control subsystem 146 may include various elements, such as an engine power output subsystem, a brake unit, a navigation unit, a steering system, and an autonomous control unit. The engine power output may control the operation of the engine, including the torque produced or the horsepower provided, and provide control over the gear selection of the transmission. The brake unit may include any combination of mechanisms configured to decelerate the autonomous vehicle 105. The brake unit may use friction to slow the wheels in a standard manner. The brake unit may include an anti-lock brake system (ABS), which may prevent the brakes from locking when the brakes are applied. The navigation unit may be any system configured to determine the driving path or route of the autonomous vehicle 105. The navigation unit may be additionally configured to dynamically update the driving path while the autonomous vehicle 105 is operating. In some embodiments, the navigation unit may be configured to combine data from a GPS device and one or more predetermined maps to determine the driving path of the autonomous vehicle 105. The steering system may represent any combination of mechanisms operable to adjust the heading of the autonomous vehicle 105 in either an autonomous mode or a driver-controlled mode.

[0064] The autonomous control unit may represent a control system configured to identify, evaluate, and avoid or otherwise negotiate potential obstacles in the environment of the autonomous vehicle 105. In general, the autonomous control unit may be configured to control the operation of the autonomous vehicle 105 without a driver, or to provide driver assistance in controlling the autonomous vehicle 105. In some embodiments, the autonomous control unit may be configured to combine data from a GPS device, RADAR, LiDAR (e.g., LIDAR), camera, and / or other vehicle subsystems to determine a driving path or trajectory for the autonomous vehicle 105. Autonomous control may activate systems that the autonomous vehicle 105 has that are not present in conventional vehicles, including those systems that may allow the autonomous vehicle to communicate with surrounding drivers or signal surrounding vehicles or drivers to safely operate the autonomous vehicle.

[0065] The onboard control computer 150 (which may be referred to as a VCU) includes a vehicle subsystem interface 160, a driving operation module 168, one or more processors 170, a compliance module 166, a memory 175, and a network communication subsystem 178. The onboard control computer 150 controls many, if not all, operations of the autonomous vehicle 105 in response to information from various vehicle subsystems 140. One or more processors 170 perform operations that allow the system to determine the health of the autonomous vehicle (such as whether the autonomous vehicle is faulty or encounters a situation that requires service or a deviation from normal operation and gives instructions). Data from the vehicle sensor subsystem 144 is provided to the VCU 150 so that the state of the autonomous vehicle can be determined. The compliance module 166 determines what actions the autonomous vehicle 105 should take to operate in accordance with applicable (e.g., local) regulations. Data from other vehicle sensor subsystems 144 can be provided to the compliance module 166 so that the best course of action can be appropriately determined and executed given the state of the autonomous vehicle. Alternatively or additionally, the compliance module 166 may determine a course of action in conjunction with another operation or control module, such as the driving operation module 168 .

[0066] The memory 175 may also contain additional instructions, including instructions for sending data to, receiving data from, interacting with, or controlling one or more of the vehicle drive subsystem 142, the vehicle sensor subsystem 144, and the vehicle control subsystem 146 including the autonomous control system. The in-vehicle control computer (VCU) 150 can control the functions of the autonomous vehicle 105 based on inputs received from various vehicle subsystems (e.g., the vehicle drive subsystem 142, the vehicle sensor subsystem 144, and the vehicle control subsystem 146). In addition, the VCU 150 can send information to the vehicle control subsystem 146 to direct the trajectory, speed, signaling behavior, etc. of the autonomous vehicle 105. The autonomous control vehicle control subsystem can receive an action plan to be taken from the compliance module 166 of the VCU 150, and thus relay instructions to other subsystems to execute the action plan. In Sections III to XXXI below, this patent document describes an autonomous vehicle or system performing certain functions or operations. Figure 16-Figure 22 The functions and / or operations described in the specification may be performed by the compliance module 166 and / or the driving operation module 168 .

[0067] Figure 2 A flow chart for safely operating an autonomous vehicle (AV) based on the health of the autonomous vehicle and / or the surrounding environment is shown. Although the figure depicts the functional steps in a particular order for illustrative purposes, the process is not limited to any particular order or arrangement of steps. Those skilled in the relevant art will understand that the various steps depicted in the figure may be omitted, rearranged, combined, and / or adapted in various ways.

[0068] like Figure 2As shown, in step 205, the vehicle sensor subsystem 144 receives visual signals, auditory signals, or audiovisual signals indicating the environmental conditions of the autonomous vehicle, and receives vehicle health or sensor activity data. As in step 210, these visual and / or auditory signal data are sent from the vehicle sensor subsystem 144 to the vehicle control computer system (VCU) 150. In step 215, any one of the driving operation module and the compliance module receives the data sent from the vehicle sensor subsystem. Then, in step 220, one or both of these modules determine whether the current state of the autonomous vehicle allows it to proceed in a usual manner, or whether the autonomous vehicle needs to change its course to prevent damage or injury or allow for maintenance. Information indicating the need to change the course of the autonomous vehicle may include: an indication of a sensor failure; an indication of a failure in the engine, brakes, or other components that may be required for the operation of the autonomous vehicle; determining visual instructions from officials, such as flares, cones, or signs; determining the presence of official personnel on the road; determining that a law enforcement vehicle is approaching the autonomous vehicle on the road, including determining from which direction; and determining that a law enforcement vehicle or a first response vehicle is moving away from the autonomous vehicle or on a road separate from the autonomous vehicle. In step 225, this information indicating that a change in the autonomous vehicle's course of action or driving-related operation is required can be used by the compliance module to formulate a new course of action to be taken that takes into account the health of the autonomous vehicle and the surrounding environment. The course of action to be taken may include slowing down, stopping, moving into the shoulder of the road, changing route, changing lanes while maintaining the same general route, etc. The course of action to be taken may include initiating communication with any supervisory system or human-computer interaction system present on the autonomous vehicle. In step 230, the course of action to be taken can then be sent from the VCU 150 to the autonomous control system. In step 235, the vehicle control subsystem 146 then causes the autonomous vehicle 105 to operate in accordance with the course of action to be taken received from the VCU 150.

[0069] It should be understood that the specific order or hierarchy of steps in the process disclosed herein is an example of an exemplary method. Based on design preferences, it should be understood that the specific order or hierarchy of steps in the process can be rearranged while remaining within the scope of the present disclosure. The attached method claims present the elements of the various steps in a sample order and are not meant to be limited to the specific order or hierarchy presented.

[0070] II. Autonomous Truck Supervision System

[0071] Figure 3A system 300 is shown that includes one or more autonomous vehicles 105, a control center or supervisory system 350 with an operator 355, and an interface 362 for third party 360 interaction. The operator 355 may also be referred to as a remote center operator (RCO). Communications between the autonomous vehicles 105, the supervisory system 350, and the user interface 362 occur over a network 370. In some cases where not all autonomous vehicles 105 in a fleet are capable of communicating with the supervisory system 350, the autonomous vehicles 105 may communicate with each other over the network 370 or directly. As described in reference Figure 1 As described above, the VCU 150 of each autonomous vehicle 105 may include a module 178 for network communications.

[0072] The self-driving trucks may communicate with the supervision system. The supervision system may be used for many purposes, including: tracking the progress of one or more self-driving vehicles (e.g., self-driving trucks); tracking the progress of a fleet of self-driving vehicles; sending maneuvering instructions to one or more self-driving vehicles; monitoring the health of the self-driving vehicle(s); monitoring the cargo status of each self-driving vehicle in contact with the supervision system; facilitating communications between third parties (e.g., law enforcement, customers whose cargo is being carried) and each or specific self-driving vehicle; allowing tracking of specific self-driving trucks in communication with the supervision system (e.g., third parties tracking a subset of vehicles in a fleet); arranging maintenance services for self-driving vehicles (e.g., oil changes, refueling, maintaining levels of other fluids); warning affected self-driving vehicles of traffic or weather changes that may adversely affect routes or delivery plans; pushing over-the-air updates to self-driving trucks to keep all components up to date; and other purposes or functions that improve the safety of self-driving vehicles, their cargo, and their surroundings. The supervision system may also determine performance parameters of the autonomous vehicle or autonomous truck, including any of the following: data logging frequency, compression rate, location, data type; communication prioritization; frequency of servicing the autonomous vehicle (e.g., mileage between services); time to perform minimum risk condition (MRC) maneuvers while monitoring the vehicle's progress during the maneuver; time to hand over control of the autonomous vehicle to a human driver (e.g., at a destination marshalling yard); ensuring that the autonomous vehicle passes pre-trip inspection; ensuring that the autonomous vehicle executes or complies with legal requirements at checkpoints and weigh stations; ensuring that the autonomous vehicle executes or complies with instructions from humans at roadblocks, crosswalks, intersections, construction, or accident sites; and so on.

[0073] Some of the functions performed by the supervisory system or command center include the ability to relay over-the-air real-time weather updates to the autonomous vehicles in the monitored fleet. Over-the-air weather updates may be pushed to all autonomous vehicles in the fleet, or may be pushed only to autonomous vehicles currently on a mission to deliver cargo. Alternatively or additionally, priority in pushing or sending over-the-air weather reports may be given to fleet vehicles that are currently on a trajectory or route heading toward a severe weather event or within a predetermined radius of a severe weather event.

[0074] Another function that may be covered by the functions performed by the supervisory system or command center is sending trailer metadata to the computing unit (VCU) of the autonomous vehicle before the cargo transport mission begins. The trailer metadata may include the type of cargo being sent, the weight of the cargo, temperature thresholds for the cargo (e.g., the temperature inside the trailer should not drop below or rise above a predetermined temperature), time sensitivity, acceleration / deceleration sensitivity (e.g., due to the fragility of the cargo, jerky motion may be harmful), trailer weight distribution along the length of the trailer, cargo packaging or stacking within the trailer, etc.

[0075] To allow communication between autonomous vehicles in a fleet and a supervisory system or command center, each autonomous vehicle may be equipped with a communication gateway. The communication gateway may have the ability to perform any of the following operations: allow communication from the autonomous vehicle to the supervisory system (e.g., V2C) and from the supervisory system to the autonomous vehicle (C2V); allow communication from the autonomous vehicle to the autonomous vehicle within the fleet (V2V); send the availability or status of the communication gateway; confirm the receipt of communications; ensure safety around remote commands between the autonomous vehicle and the supervisory system; reliably communicate the location of the autonomous vehicle at set time intervals; enable the supervisory system to ping the location and vehicle health of the autonomous vehicle; allow various sensor data to flow directly to the command or supervisory system; allow automatic alerts between the autonomous vehicle and the supervisory system; comply with the ISO 21434 standard; and so on.

[0076] The supervisory system or command center may be operated by one or more persons, also referred to as operators or remote center operators (RCOs). The operators may set thresholds for autonomous vehicle health parameters so that when the autonomous vehicle meets or exceeds the thresholds, preventive measures may be taken. Examples of vehicle health parameters for which thresholds may be established by the operator may include any of the following: fuel level; oil level; mileage since last maintenance; low tire pressure detected; washer fluid level; brake fluid level; responsiveness of steering and brake subsystems; diesel exhaust fluid (DEF) level; communication capabilities (e.g., lack of responsiveness); positioning sensor capabilities (e.g., GPS, IMU failure); impact detection (e.g., vehicle collision); perception sensor capabilities (e.g., camera, LIDAR, radar, microphone array failure); computing resource capabilities (e.g., VCU or ECU failure or lack of responsiveness, temperature anomalies in computing units); angle between tractor and trailer in towing situations (e.g., semi-trailer, 18-wheeler, or semi-trailer); unauthorized entry of a living entity (e.g., a person or animal) into the interior of an autonomous truck; etc. Preventive measures may include performing a minimum risk condition (MRC) maneuver, seeking maintenance, or leaving the road or other such route changes that are not too strenuous for the autonomous vehicle. An autonomous vehicle whose system health data meets or exceeds a threshold set at the supervisory system or by an operator may receive instructions automatically sent from the supervisory system to perform preventive measures.

[0077] Operators may be made aware of situations that affect one or more autonomous vehicles in communication with or monitored by a supervisory system, while the affected autonomous vehicle(s) may not be aware of these situations. These situations may include: irregular or sudden changes in traffic flow (e.g., traffic jams or accidents); sudden changes in weather; sudden changes in visibility; emergency conditions (e.g., fires, collapses, bridge cracks); power outages affecting traffic lights; unexpected road works; large or obscure road debris (e.g., objects that cannot be identified by the autonomous vehicle); law enforcement activities on the road (e.g., car chases or road clearing activities); and the like. The supervisory system operator may be made aware of these types of situations that may not be detected by the autonomous vehicle through traffic reports, law enforcement communications, data from other vehicles in communication with the supervisory system, reports from drivers of other vehicles in the area, and similar distributed information venues. The autonomous vehicle may not be able to detect these situations due to limitations of the sensor system or lack of access to information distribution means (e.g., no direct communication with weather agencies). The operator at the supervisory system may push this information to the affected autonomous vehicles in communication with the supervisory system. The affected autonomous vehicles may continue to change their routes, trajectories, or speeds in response to information pushed from the supervisory system. In some cases, information received by the supervisory system may trigger a threshold condition indicating that an MRC (minimum risk condition) maneuver is approved; alternatively or additionally, an operator may assess the situation and determine that the affected autonomous vehicles should perform an MRC maneuver, and then send such instructions to the affected vehicles. In these cases, each autonomous vehicle that receives information or instructions from the supervisory system or supervisory system operator uses its onboard computing unit (e.g., VCU) to determine how to proceed safely, including performing an MRC maneuver including pulling over or stopping.

[0078] Other interactions that a remote center operator (RCO) may have with an autonomous vehicle or fleet of autonomous vehicles include any of the following: pre-planned event avoidance; real-time route information updates; real-time route feedback; trailer hitch status; first responder communication request processing; notification of aggressive surrounding vehicle(s); identification of construction zone changes; autonomous vehicle status regarding its operational design domain (ODD), such as alerting the RCO when the autonomous vehicle approaches or enters a state in the ODD; RCO notification when the autonomous vehicle is within a threshold distance of a toll booth, and appropriate instructions / communications with the autonomous vehicle or toll collection agency may be sent to allow the autonomous vehicle to bypass the toll; RCO notification when the autonomous vehicle bypasses a toll; RCO notification when the autonomous vehicle is within a threshold distance of a weigh station, and appropriate instructions / communications with the autonomous vehicle or appropriate agency may be sent to allow the autonomous vehicle to bypass the weigh station; RCO notification when the autonomous vehicle bypasses a weigh station; notification from the RCO to the autonomous vehicle regarding dispatch or refueling or maintenance needs; RCO authorization of third party access to the autonomous vehicle cab; RCO startup / restart of the autonomous driving system on the vehicle The invention also provides for the ability of an autonomous vehicle to provide a reliable and secure environment for the autonomous vehicle and to provide a reliable and secure environment for the autonomous vehicle. The invention also provides for the ability of an autonomous vehicle to provide a reliable and secure environment for the autonomous vehicle and ...

[0079] The supervisory system or command center may allow third parties to interact with the supervisory system operator, with the self-driving truck, or with the human-machine system operator and the self-driving truck. The third party may be a customer whose goods are being transported, a law enforcement agency or emergency service provider, or someone assisting the self-driving truck when service is needed. In its interactions with third parties, the supervisory system may identify different levels of access so that a customer concerned about the timing or progress of a shipment may only be allowed to view status updates of the self-driving truck, or may be able to view the status and provide input to the supervisory system about what parameters to prioritize (e.g., speed, economy, maintaining the originally planned route). By providing input to the supervisory system about parameter prioritization, the customer can influence the route and / or operating parameters of the self-driving truck.

[0080] The actions that the autonomous vehicles (particularly autonomous trucks) described herein may be configured to perform in order to safely traverse the process while complying with applicable rules, laws, and regulations may include those actions that are successfully completed by a human-driven autonomous truck. These actions or maneuvers may be described as features of the truck because they may be executable programming stored on the VCU 150 (vehicle control computer unit). These actions or features may include actions or features related to reactions to detection of certain types of conditions or objects, such as: appropriate movement on a hill; appropriate movement on a curve, appropriate movement at a highway exit; appropriate movement or action in response to: detection of one or more stopped vehicles, detection of one or more vehicles in an emergency lane, detection of an emergency vehicle with flashing lights that may be approaching the autonomous vehicle; movement in response to detection of one or more large vehicles approaching, adjacent to, or about to be adjacent to the autonomous vehicle; movement or action in response to pedestrians, cyclists, etc. after identification and classification of these persons; movement or action in response to curved or inclined portions of the road; and / or movement in response to encountering intersections in response to identifying on-ramps and off-ramps on a highway or freeway; merging into traffic in an adjacent lane or traffic area; detecting the need to clean one or more sensors and cleaning appropriate sensors; etc. Other features of the autonomous truck may include those actions or features required for any type of maneuver, including those required to complete the reactive features or actions listed above.

[0081] Supporting features may include: safe lane changes; operating turn signals on the autonomous truck to alert other drivers of intended changes in motion; offsetting the autonomous truck in its lane (e.g., away from the center of the lane to accommodate the motion or size of adjacent vehicles or approaching objects); the ability to maintain an appropriate following distance; the ability to turn right and left using appropriate signals and actions, etc. Supporting features may also include: the ability to navigate roundabouts; the ability to illuminate appropriately with onboard lights based on ambient light and the need to comply with local laws; the minimum amount of deceleration required to apply any given maneuver; determining position at all times; dynamic vehicle control adapted to trailer load distribution, excluding wheel adjustments; starting (reaching target speed), accelerating, stopping, and yielding; operating on uneven roads; entering a minimum risk condition (MRC) on the shoulder; obtaining local laws and regulations based on position along the route; operating on asphalt, concrete, mixed gentle slopes, scratched roads, and gravel; the ability to operate based on light metering lights / signals at entry ramps; operating on roads with a width up to a predetermined width; being able to stop at crosswalks with sufficient stopping distance; navigating two-way left turn lanes; operating on roads with entry and exit ramps; communicating with other drivers using the vehicle horn; etc. One or more of the features and / or one or more of the supporting features described in this patent document may be combined and may be performed by an onboard control computer in an autonomous truck.

[0082] In some embodiments, actions or features may be considered supporting features and may include: speed control; the ability to maintain a straight path; etc. These supporting features, as well as the reactive features listed above, may include controlling or modifying steering, engine power output, braking, or other vehicle control subsystems 146. The reactive features and supporting features listed above are discussed in more detail below.

[0083] III. Curves and / or general bends

[0084] Self-driving trucks are able to safely navigate curved roads. Such navigation includes the ability to detect whether a highway or a portion of a road is curved. Various types of data, including mapping, navigation, and perception data (e.g., images, radar data, LIDAR data), can be used to identify curved sections of roads. Once the curved portion of the road or highway is identified, the self-driving vehicle (e.g., a self-driving truck) can be controlled to maneuver appropriately. The self-driving vehicle can be controlled to stay within the boundaries of its current lane, making appropriate speed and steering adjustments to stay in its lane. The self-driving vehicle control system can change the positioning of the self-driving vehicle in its lane taking into account surrounding vehicles or objects, that is, adjusting lane offset, which will be described in more detail below. There can also be a threshold amount, and the self-driving vehicle or truck can be allowed to move the threshold amount to adapt to the surrounding vehicles and bends in the road. If the threshold amount is exceeded, the self-driving truck can perform alternative maneuvers, such as slowing down, temporarily leaving the road, or changing the route to travel safely.

[0085] III.(a) Description of the curve

[0086] A curve may be defined as a continuous road segment that is not interrupted by a traffic intersection having a minimum radius of curvature between predetermined thresholds, such as 1250 m, including 1200 m, 1150 m and 1100 m.

[0087] In some embodiments, to obtain less than 1 m / s in bending 2 To provide a higher level of comfort and stability in lateral acceleration, the minimum radius of curvature that the autonomous vehicle can take without decelerating at a maximum operating speed of 75 mph (33.5 m / s) is calculated to be within predetermined thresholds, including a minimum radius of curvature of 1112 m, 1124 m, or 1136 m.

[0088] III.(b) draw

[0089] The autonomous vehicle may map curved road segments for navigation purposes. The autonomous vehicle may have the curved road segments and associated semantics included in the map for navigation purposes. The associated semantics may generally include a vocabulary for identifying features associated with the curve.

[0090] III.(c) Minimum bend avoidance

[0091] The autonomous vehicle may avoid curves where the minimum radius of curvature is less than a threshold or a predetermined distance. The threshold for the minimum radius of curvature of a curve may be based on the truck configuration (e.g., the length of the trailer, the position of the trailer's front wheels, the position of the fifth wheel hook). The threshold for the minimum radius of curvature of a road on which an autonomous vehicle (i.e., an autonomous truck) can safely operate may be in the range of 15m to 20m, such as 17m to 19m, including 18m. Similarly, the minimum curvature value threshold may be in the range of 0.40rad / m to 0.70rad / m, such as 0.45rad / m to 0.65rad / m or 0.50rad / m to 0.60rad / m, including 0.055rad / m.

[0092] III.(d) Smooth acceleration and deceleration

[0093] In addition to evasive maneuvers, the automated vehicle may use a predetermined threshold amount or predetermined speed (m / s 2 ) to prevent jolts that could upset the balance of the autonomous vehicle in a bend. The predetermined rate can be from 0.5 m / s 2 Up to 10m / s 2 In the range of 1m / s 2 Up to 8m / s 2 , and includes 2m / s 2 Up to 6m / s 2 .

[0094] III.(e) Curve Sign Recognition

[0095] The autonomous vehicle is able to identify a curve based on a sign. The sign may be identified based on the color of the sign, the shape of the sign, and an illustration on the sign. Alternatively or additionally, the sign may include text indicating any of a curve ahead, a speed limit, a distance to a curve in the road ahead, reduced visibility due to a curve ahead, and a change in passing convention due to a curve in the road ahead.

[0096] III.(f) Curve deviation

[0097] When driving on a curved highway section with a minimum radius of curvature greater than or equal to a predetermined threshold distance, the autonomous driving vehicle may deviate from the center of the lane toward the convex side of the curve (or the side that curves outward) up to the predetermined threshold distance. The predetermined threshold distance may be in a value range such as 0.25 to 1.5 m, 0.25 to 1.25 m (including 0.30 to 1.0 m).

[0098] III.(g) Visibility Time to Collision (TTC)

[0099] When encountering a curve, the autonomous vehicle may decelerate to maintain TTC for visibility for at least a predetermined amount of time. For example, when encountering a curve, the autonomous vehicle may decelerate to maintain TTC for forward visibility of a front NPC or object in the lane for at least a predetermined number of seconds. The time to collision (TTC) for the predetermined number of seconds of forward visibility may fall within a range of values ​​such as between 2s and 8s, such as between 2s and 6s, including between 3s and 5s.

[0100] III.(h) Avoidance of superelevation on curves

[0101] The automated vehicle may avoid curves where the superelevation (or the upward angle of the curve relative to a flat surface) exceeds a predetermined threshold amount or a predetermined percentage. The predetermined percentage of superelevation may be in the range between 5% and 15%, such as between 7% and 12%, including between 8% and 10%.

[0102] According to the U.S. Department of Transportation, superelevation is the curl of the pavement when approaching and passing a horizontal curve. This type of pavement curl is intended to assist the driver by counteracting the lateral acceleration created by tracking the curve.

[0103] III.(i) Lane change avoidance

[0104] When on a curve, the autonomous vehicle may avoid all types of efficiency lane changes and lower priority lane changes. An efficiency lane change is a lane change where efficiency is the reason for the lane change. For example, a lane change intended because staying in the current lane will result in an unexpected exit from the highway onto a local road within a predetermined distance (e.g., 1100m, 1200m, 1300m, 1400m) is an efficiency lane change. This type of lane change will avoid having to exit the highway and use local roads to get back on track. Other types of efficiency lane changes may include: an intended lane change due to an upcoming planned exit between 1600m (1 mile) and 3200m from the autonomous vehicle; an intended lane change because staying in the current lane will result in an unexpected exit from the highway within a predetermined distance greater than 1600m (1 mile) and less than 3200m; and an intended lane change to bypass a slow vehicle traveling within a predetermined range below the ambient speed (e.g., between 5mph and 25mph below the ambient speed, 10mph and 20mph below the ambient speed).

[0105] III.(j) Off-track area

[0106] Unless changing lanes or drifting, the autonomous vehicle can choose a trajectory so that the off-track region is within the boundaries of the lane lines when in a curve. In the vehicle field, off-track occurs when the vehicle makes a turn and the rear wheels do not follow the same path as the front wheels. The off-track region is the area swept by the rear wheels of the autonomous vehicle when traversing a curve.

[0107] III.(k) Curve Detection

[0108] The autonomous vehicle can detect the curvature of the road using onboard sensors located on or in the autonomous vehicle. The onboard sensors used to detect the road configuration may include visual cameras, LIDAR, radar, GPS or other positioning systems, time-of-flight cameras, ultrasonic sensors, etc.

[0109] III.(l) Speed ​​reduction

[0110] The autonomous vehicle may reduce speed when driving on a curve to prevent the trailer from tipping, swaying, or sliding. Sensors that may detect tipping, swaying, or sliding of the autonomous vehicle or the trailer portion of the autonomous vehicle may include any of one or more inertial measurement units (IMUs), one or more gyroscopes, onboard cameras, one or more ultrasonic sensors, onboard LIDAR data, detectors for irregular changes in wheel slip on the road surface, sensors for detecting changes in steering angle while maintaining a planned trajectory (i.e., this may be an indicator of changes in the distribution of loads or other forces between the axles / wheels of the vehicle), etc.

[0111] The autonomous vehicle may take into account superelevation, road curvature, road traction conditions, prevailing weather conditions, visibility conditions, and the weight and center of gravity of the autonomous vehicle and trailer.

[0112] III.(m) Do not cross lane boundaries

[0113] The automated vehicle must not deviate from the center of the lane so that any part of the truck (including the rearview mirrors) crosses the nearest edge of the lane boundary, except when conducting an evasive maneuver or when a planned deviation is required.

[0114] III.(n) Off-center threshold of nearby vehicles

[0115] When there is a neighboring vehicle parallel to the autonomous vehicle on a curve, the autonomous vehicle must not deviate from the center of the lane so that any part of the truck (tractor or trailer, including rearview mirrors) comes within a predetermined distance from the lane boundary that intersects the autonomous vehicle and nearby vehicles (e.g., NPC vehicles). The predetermined distance from the lane boundary can be in the range of 0.1m to 0.5m, such as 0.15m to 0.4m, including 0.2m to 0.3m. The distance from the lane boundary should be measured from the edge of the boundary closest to the autonomous vehicle. This constraint only applies to lanes equal to or greater than the U.S. standard highway lane width, for example 3.66m (12 feet). If the on-board sensors detect that there is a certain degree of overlap, the vehicle can be considered parallel to the autonomous vehicle.

[0116] For example, in one embodiment, the width of the autonomous vehicle including the rearview mirror and side sensor assembly may be between 2.6m and 2.7m. The positioning sensors and systems of the autonomous vehicle may include an error associated with the distance from the center of the lane. This may allow a predetermined distance from the lane boundary to be met.

[0117] III.(o) Deviation of nearby vehicles

[0118] When there is a large adjacent vehicle nearby, the autonomous vehicle can make a critical safety offset to move away from that vehicle.

[0119] When there is a neighboring non-large vehicle nearby, the autonomous vehicle can make a non-critical safety offset to move away from the vehicle.

[0120] Large vehicles may include vehicles greater than 7m in length or oversized vehicles. Consumer vehicles without a trailer attached may not be defined as large vehicles.

[0121] A critical safety excursion is when the autonomous vehicle moves away from a hazard within its current lane. The autonomous vehicle may relax lane boundaries as needed. Situations where a critical safety excursion may be required are at least the following: when driving parallel to a noncompliant vehicle and the autonomous vehicle cannot change lanes or is in the process of changing lanes; when driving parallel to an oversized vehicle; when driving parallel to a large vehicle on a curve; when driving parallel to a large vehicle on a narrow lane; and when driving parallel to an ELV in the presence of pedestrians and the autonomous vehicle cannot change lanes or is in the process of changing lanes.

[0122] The preferred behavior in the above situation may be to perform a different maneuver, such as a lane change. The deviation will only apply when the lane change or other preferred behavior cannot be performed or the automated driving vehicle is in the process of performing it.

[0123] Fig.18 An example flow chart of an autonomous driving operation performed by a vehicle to operate on a curved area of ​​a road is shown. Operation 1802 includes determining, by a computer located in the autonomous vehicle, that a lane of a road on which the autonomous vehicle is operating includes a curved portion having a minimum radius greater than or equal to a predetermined threshold. Operation 1804 includes operating the autonomous vehicle operating on the curved portion to move toward a side of the lane and away from a center of the lane as the autonomous vehicle drives through the curved portion of the road, wherein the side of the lane is a side that curves outward, and wherein the autonomous vehicle is caused to move toward the side of the lane until a predetermined threshold distance from the center of the lane. In some embodiments, operating the autonomous vehicle to move toward one side of the lane in operation 1804 includes sending instructions to one or more devices (e.g., one or more motors) in a steering system of the autonomous vehicle to cause the autonomous vehicle to steer.

[0124] In some embodiments, the method further includes: operating the autonomous vehicle to avoid driving on the additional curved portion of the road in response to determining that the additional curved portion of the road has a superelevation greater than a predetermined threshold amount, wherein the superelevation describes an upward angle formed by the additional curved portion tilting upward relative to the flat surface. In some embodiments, the predetermined threshold amount is a value between 5% and 15%. In some embodiments, the method further includes operating the autonomous vehicle to accelerate or decelerate in the curved portion of the road at a rate less than or equal to a predetermined rate. In some embodiments, the method further includes: operating the autonomous vehicle to reduce the speed of the autonomous vehicle in response to determining that a time to collision (TTC) value when the autonomous vehicle is operating in the curved portion is greater than a predetermined amount of time, wherein the TTC value indicates the amount of time of visibility provided by one or more cameras on the autonomous vehicle.

[0125] IV. Emergency lane vehicles on highways or roads

[0126] An autonomous vehicle (e.g., an autonomous truck) is able to maneuver appropriately when encountering one or more vehicles in an emergency lane on a road the truck is traversing. The feature includes the ability to identify the presence of at least one vehicle in an emergency lane (e.g., a shoulder lane). The ability to identify the presence of a vehicle in an emergency lane includes identifying the emergency lane, identifying the type of vehicle in the emergency lane, identifying the likelihood that the vehicle will enter or exit traffic in the emergency lane, and identifying the location of one or more vehicles in the emergency lane.

[0127] The appropriate action for any vehicle on the highway overtaking on the shoulder or emergency lane may depend on many factors, including any one or more of the following: surrounding traffic, the desired trajectory of the moving vehicle, the presence or proximity of emergency responders or law enforcement, the presence of pedestrians adjacent to the pulled-over vehicle, road debris, and / or local regulations.

[0128] After detecting the presence of one or more vehicles stopped on the shoulder or in the emergency lane, the autonomous truck may do any of the following if deemed appropriate: slow down in the current lane; move in the current lane to accommodate the stopped vehicle(s); move in the current lane to accommodate other vehicles moving to avoid the stopped vehicle(s); change lanes to create more distance between the autonomous vehicle and the stopped vehicle(s); leave the road.

[0129] In some embodiments, when an emergency vehicle is detected slowing down in a shoulder or emergency lane adjacent to the autonomous truck, the autonomous truck may slow down by a predetermined amount when the posted speed limit is 25 mph or greater (e.g., at least 20 mph when the posted speed limit is 25 mph or greater). Alternatively, if the posted speed limit is less than 25 mph, the autonomous truck traveling in a lane adjacent to the emergency lane where the emergency vehicle is parked may slow down to 5 mph.

[0130] In some cases, an emergency vehicle or other vehicle stopped in the shoulder or emergency lane may begin to move into the main traffic lane. The autonomous truck can accommodate the emergency vehicle by adjusting its speed or performing other adaptive maneuvers. This process of adapting to one or more vehicles moving from the shoulder to the main traffic lane can include considering local regulations and calculating which maneuver is safest and most feasible to perform, as well as identifying vehicles in the shoulder that want to enter traffic. Conversely, it may be necessary to accommodate emergency vehicles or other vehicles that need to enter the shoulder or emergency lane. Part of this adaptation of the autonomous truck is identifying vehicles that want to move to the shoulder, and moving in a manner that allows additional vehicles to reach the shoulder, taking into account local regulations and general safety, assuming the additional vehicles have enough power to do so.

[0131] IV.(a) Description of Highway Emergency Lane Vehicle (ELV)

[0132] An Emergency Lane Vehicle (ELV) is any type of vehicle that is outside the travel lane boundary and within a predetermined number of meters (e.g., 7m, 7.25m, 7.5m, 7.75m, 8m) from the nearest point of the NPC to the nearest outer travel lane boundary on the highway, including within paved / unpaved shoulders, emergency lanes, or lane dividers.

[0133] Traveling lane boundary means the edge line pavement marking that delineates the right or left edge of the roadway. Through FWHAMUTCD Sections 3B06 and 3B07, right edge line pavement markings will consist of normal solid white lines, and left edge line pavement markings (on divided highways or one-way streets) will consist of normal solid yellow lines; in addition, (1) expressways, (2) freeways, and (3) rural arterial roads with a travel lane of 20 feet or more and an average daily traffic volume of 6,000 vehicles or more must include edge line markings.

[0134] IV.(b) Transition from NPC to ELV

[0135] The autonomous vehicle is able to predict when an NPC will transition to become an ELV. An NPC (non-player character) can be a manually operated vehicle or another autonomous vehicle that is not controlled by the autonomous driving system of the autonomous vehicle discussed herein (sometimes referred to as an Ego). The autonomous vehicle can use onboard sensors to detect NPC characteristics or behaviors that indicate a transition from an NPC to an ELV. NPC characteristics or behaviors that may indicate a transition to an ELV may include any of the following: use of a turn signal indicating a direction toward the shoulder of the road; a change in the offset or trajectory of the NPC; activation of the NPC's hazard lights; activation of flashing lights indicating an emergency vehicle, rescue vehicle, or law enforcement vehicle; changes in the speed of the NPC; and direct communication from the NPC to the autonomous vehicle (e.g., Ego) indicating the NPC's intention to move to the emergency lane or shoulder of the road; and the like.

[0136] In some embodiments, if the NPC is decelerating and signaling its intent to turn into the shoulder of the road (or lane divider, etc.), the automated driving vehicle may determine that the NPC is transitioning to the ELV.

[0137] IV.(c) NPC to ELV transition - behavior

[0138] If the vehicle is transitioning to the ELV, the autonomous vehicle may change lanes to move away from the adjacent lane of the shoulder. If the autonomous vehicle is following a vehicle in the adjacent lane of the shoulder and cannot change lanes, the autonomous vehicle may slow down and match the speed of that vehicle until it moves out of the autonomous vehicle's lane.

[0139] IV.(d) NPC to ELV transition - Behavior - Do not change lanes

[0140] Automated vehicles must not change lanes into lanes adjacent to the ELV destination area that is transitioning.

[0141] IV.(e) Late detection of ELV

[0142] Timely detection of vehicles on the shoulder or in the diverging angle area is very important to maintain the safety of the autonomous vehicle. If the ELV is obscured from the field of view of the autonomous vehicle or is detected too late (for example, high curvature area, diverging angle area), and the system cannot meet the predetermined reaction distance requirements, the system can react immediately as soon as the ELV is detected. These behaviors remain the same as described in this section. In this scenario, when the front bumper of the autonomous vehicle is a predetermined distance (for example, 2.5m, 3m, 3.5m, 4m) in front of (or behind) the rear bumper of the ELV, the autonomous vehicle can at least complete the deceleration.

[0143] The predetermined reaction distance requirements are as follows. When an ELV is encountered in an area directly adjacent to the lane currently being traveled, the autonomous vehicle may change lanes to move away from the ELV in order to overtake. The autonomous vehicle may be designed to change lanes as soon as the ELV is detected and begin to react no later than a predetermined distance (e.g., 125m, 140m, 152m (500 feet), 175m) before reaching the ELV. In some cases, the autonomous vehicle may not be able to change lanes to move away from the ELV. When an ELV is encountered in an area adjacent to the lane currently being crossed and it is not possible to change lanes to overtake, the autonomous vehicle may begin to slow down and offset at a predetermined distance before reaching the ELV (such as no later than 125m, 140m, or 152m (500 feet) before reaching the ELV) to maintain a safe speed and lateral distance.

[0144] IV.(f) ELV approaches the driving lane

[0145] When encountering an ELV within a predetermined distance (e.g., 1.0m, 1.3m, 1.5m) (laterally) of the autonomous vehicle, the autonomous vehicle may perform a full lane change or lane deviation to at least maintain the distance (e.g., 1.0m lateral distance, 1.3m lateral distance) from the ELV. In this scenario, it is permissible for the autonomous vehicle to cross the following lane boundaries to perform a full lane change or avoid an accident.

[0146]

[0147]

[0148] IV.(g) Lane change preference

[0149] When an ELV is encountered in an area directly adjacent to the current lane, the autonomous vehicle may change lanes to move away from the ELV in order to overtake. The autonomous vehicle may be designed to change lanes as soon as the ELV is detected and begin reacting no later than a predetermined number of meters (e.g., 125 meters (m), 130m, 140m, 150m, 152m, 155m, 160m) before reaching the ELV.

[0150] IV.(h) Roadside emergency vehicles

[0151] When encountering a roadside emergency vehicle, the autonomous system may change lanes in the following manner. When encountering an ELV in an area directly adjacent to the lane currently being traveled, the autonomous vehicle may change lanes away from the ELV in order to overtake. The autonomous vehicle may be designed to change lanes as soon as the ELV is detected and begin reacting no later than a predetermined distance (e.g., 125m, 140m, 152m (500 feet), 175m) before reaching the ELV.

[0152] If changing lanes is not possible, the autonomous vehicle may slow down in accordance with the laws of the state in which the autonomous vehicle is operating, which may be defined by the geofence as shown in the table below.

[0153]

[0154] The strictest laws currently regarding roadside emergency vehicles come from Texas statutes.

[0155] IV.(i) Slowdown and Deviation Strategy

[0156] When an emergency lane vehicle (ELV) is encountered in an area adjacent to the autonomous vehicle's current lane and the autonomous vehicle is unable to change lanes to overtake, the autonomous vehicle may begin to decelerate and offset to maintain a safe speed (e.g., less than or equal to the "maximum overtaking speed" shown in Tables 1 and 2 below) and lateral distance no later than a predetermined distance before reaching the ELV (e.g., 125m, 140m, or 152m (500 feet) before reaching the ELV).

[0157] The lateral distance may refer to the vertical distance between the outermost point of the autonomous vehicle and the outermost point of the ELV, unless otherwise specified.

[0158] Key parameters (based on interviews with drivers):

[0159] Table 1: Recommended maximum overtaking speeds for ELVs on highways

[0160]

[0161] Table 2: Recommended maximum overtaking speeds for ELVs on local roads

[0162] Posted speed limit Maximum overtaking speed 55mph 40 mph, or between 35 mph and 40 mph 40mph 30 mph, or between 35 mph and 40 mph 30mph 20 mph, or between 35 mph and 40 mph 20mph 10 mph, or between 5 mph and 10 mph 10mph 10 mph, or between 5 mph and 10 mph

[0163] Table 3: Preferred offset thresholds

[0164]

[0165]

[0166] IV.(j) Pedestrians near the ELV

[0167] When there are pedestrians near the ELV, the autonomous vehicle can slow down and pass at a predetermined minimum lateral distance (e.g., 3.25m, 3.5m, 3.75m, 4.0m, 4.25m). If this is not possible, the autonomous vehicle can maintain a minimum predetermined distance from the pedestrian (e.g., 0.8m, 0.92m, 0.95m, 1.0m).

[0168] IV.(k) Emergency Lane Vehicles (ELV) near the planned exit

[0169] When approaching an exit on a planned route, if there are pedestrians near an ELV (Emergency Lane Emergency Vehicle) or ELV, it is acceptable that the autonomous vehicle prioritizes slowing down without changing lanes completely in order to avoid missing the exit. In these scenarios, the autonomous vehicle can slow down. When there is an ELV near the planned exit, the deceleration of the autonomous vehicle can be performed according to Table 4 below. Table 4 lists guidance values ​​for maximum overtaking values ​​when there is an ELV on the highway, where the maximum overtaking speed is different for various posted speed limits on the highway. Alternatively or additionally, the autonomous vehicle can slow down and exceed the ELV with pedestrians nearby at a predetermined minimum lateral distance between 2.5m and 4.5m (such as 3m and 4m, such as at least 3.75m (12 feet)). When exceeding at the predetermined minimum lateral distance, the autonomous vehicle can maintain at least a predetermined amount of distance from any pedestrians present. In some such cases, the predetermined distance between pedestrians and the autonomous vehicle may be specified by law or regulation, such as 0.92m (3 feet) under Arizona law.

[0170] Table 4 Recommended maximum overtaking speeds when there is an emergency lane vehicle (ELV) on the highway

[0171]

[0172]

[0173] IV.(l) ELV near speed limit change area

[0174] When encountering an ELV in a speed limit change zone, the autonomous vehicle has three states:

[0175] 1) Autonomous vehicles are at old speeds

[0176] 2) The autonomous vehicle is changing speed to a new speed

[0177] 3) Autonomous vehicles at a new speed

[0178] The autonomous vehicle can follow the following strategies based on different ELV locations: State (1) Strategy - ELV in speed limit change zone, State (2) Strategy - Speed ​​limit increased, State (3) Strategy - Speed ​​limit decreased, State (4) Strategy - ELV in speed limit change zone

[0179] The new speed limit takes effect at the speed limit sign.

[0180] IV.(m) State (1) Strategy - ELV in Speed ​​Limit Change Area

[0181] If the ELV is within (1), the autonomous vehicle can decelerate according to the original speed limit.

[0182] IV.(n) State (2) Strategy: Autonomous vehicles near ELVs and areas with increased speed limits

[0183] An automated vehicle accelerating due to an increased posted speed limit near an emergency lane vehicle (ELV) may require different behaviors based on the location of the ELV. An ELV on a sloped section of a highway may require a first type of behavior by the automated vehicle, while an ELV on a straight portion of a highway may require a second type of behavior by the automated vehicle, and an ELV on a curved portion of a highway may require a third type of behavior by the automated vehicle.

[0184] Generally speaking, when the autonomous vehicle accelerates due to a speed limit increase, the autonomous vehicle may stop accelerating when the ELV is within a predetermined distance of the autonomous vehicle. The autonomous vehicle may overtake the ELV at a speed that does not exceed the maximum overtaking speed for the portion of the road where the center of the ELV is located. In addition, when the ELV is in a non-ramp section and the autonomous vehicle has not yet begun to accelerate, the autonomous vehicle may maintain the current posted speed and decelerate, thereby treating the ELV as if it was before the posted speed change, regardless of the position of the ELV. In this scenario, for the ELV that the autonomous vehicle is overtaking, and in the case where the posted speed limit is higher than the position where the autonomous vehicle was previously located, when the autonomous vehicle has begun to accelerate, the autonomous vehicle may stop accelerating and reduce the speed by a required amount from the maintained speed. In some embodiments, the autonomous vehicle may detect the ELV and use the lower speed limit in a table (e.g., Table 4 above) to decide how much to decelerate. For example, when the autonomous vehicle is at 67mph because it recently moved into a 70mph zone, but detects the ELV, the autonomous vehicle may use the 65mph speed limit to decelerate when it overtakes the ELV.

[0185] IV.(o) State (2) Strategy - Speed ​​Limit Reduction

[0186] If the ELV is within (2) and the speed limit is decreasing, the automated vehicle can decelerate to ensure that its deceleration profile reaches the ELV passing speed required for the ELV's actual location, even if that speed is lower than the upcoming posted speed.

[0187] The speed limit takes effect from the location of the speed limit sign.

[0188] IV. (p) State (3) Strategy - ELV in Speed ​​Limit Change Area

[0189] If the ELV is within (3), the autonomous vehicle can maintain the new passing speed constraint.

[0190] IV.(q) ELV confluence - prediction

[0191] The autonomous vehicle is able to predict when the ELV is attempting to merge back into traffic. When the autonomous vehicle predicts that the ELV is attempting to merge back into traffic, the autonomous vehicle can determine, based on the movement of the ELV, whether the merging ELV will attempt to cut in in front of or behind the autonomous vehicle.

[0192] For example, if the ELV has its turn signal on, is offset in the direction of traffic, and is accelerating slightly, then the autonomous vehicle may determine that the ELV is likely attempting to merge back into traffic. Conversely, if the garbage collection truck is in motion, but is not offset in the direction of traffic and has not engaged its signal, then the autonomous vehicle may determine that the ELV is likely not attempting to merge.

[0193] IV.(r) Deceleration Levels and Timing When the front bumper of the autonomous vehicle is 3m ahead of the rear bumper of the ELV, the autonomous vehicle may use the minimum required deceleration to achieve the predetermined travel speed and distance (engine braking or engine braking deceleration level (~-1m / s 2 ) is preferred). The predetermined travel speed and distance may comply with behaviors including that the autonomous vehicle may begin to decelerate and offset to maintain a safe speed and lateral distance no later than a predetermined distance before reaching the ELV (e.g., 125m, 140m, or 152m (500 feet) before reaching the ELV).

[0194] IV.(s)ELV Memory

[0195] The autonomous vehicle can locate each ELV and store its position until the autonomous vehicle passes each corresponding ELV. This can be achieved using data from the on-board sensor suite and a computing module or a module for identifying and tracking surrounding objects at all times.

[0196] IV. (t) Acceleration near ELV

[0197] When the autonomous vehicle is approaching an ELV in an adjacent lane and is within a predetermined distance (e.g., 125m, 140m, or 152m), the autonomous vehicle must not accelerate (even when at a slower speed), except for lane changes or evasive maneuvers (e.g., to prevent an accident or hit a pedestrian / cyclist). When the rear bumper of the autonomous vehicle (or trailer) exceeds the front bumper of the ELV by a certain distance (e.g., 2.5m, 3m, or 3.5m), the autonomous vehicle can start accelerating again.

[0198] IV.(u)ELV confluence - behavior

[0199] The maneuver of an emergency lane vehicle (ELV) merging into a lane of traffic from a shoulder or other area adjacent to a major roadway can be complex for an autonomous vehicle to engage. The following are examples of behaviors that an autonomous vehicle may exhibit or perform to increase the probability of safely merging into traffic with a detected ELV.

[0200] In a scenario where an autonomous vehicle is driving in a lane adjacent to an ELV and the ELV is moving, the autonomous vehicle may prioritize lane changes in order to minimize interactions with NPCs. The autonomous vehicle may perform lane changes as follows. When an ELV is encountered in an area directly adjacent to the current lane being crossed, the autonomous vehicle may change lanes to move away from the ELV in order to overtake. The autonomous vehicle may be designed to change lanes as soon as the ELV is detected and begin reacting no later than a predetermined number of meters (e.g., 125 meters (m), 130m, 140m, 150m, 152m, 155m, or 160m) before reaching the ELV.

[0201] When the autonomous vehicle cannot pass a complete lane between itself and the moving ELV, the autonomous vehicle may decelerate and move in its lane (i.e., shift in the lane) to safely pass the moving ELV. In this scenario, the autonomous vehicle may begin to decelerate and shift no later than a predetermined distance before reaching the ELV (e.g., 125m, 140m, or 152m (500 feet) before reaching the ELV) to maintain a safe speed and lateral distance.

[0202] The autonomous vehicle may predict whether a moving ELV will cut in. This prediction may be based on sensed motion or indications (e.g., turn signal usage, acceleration and motion toward traffic in an active lane of a highway or road, activation of a siren or other audio signal).

[0203] If the moving ELV is predicted to cut in, the autonomous vehicle may accept the cut in, otherwise the autonomous vehicle may overtake the moving ELV. The autonomous vehicle may make the decision to accept the moving ELV cut in or overtake such a vehicle based on the distance to the ELV when the ELV's intention is identified, as well as the speed, acceleration, and planned trajectory of the ELV. This decision may be governed by the threshold distance maintained between the autonomous vehicle and the moving ELV.

[0204] IV.(v) Stop and wait for ELV to be closed

[0205] If the ELV is so close that there is a possibility that the autonomous vehicle could collide (also considering the uncertainty in the trailer's position and our full offset within the lane), it is preferred that the autonomous vehicle not cross over the ELV and either change lanes completely with the lane change signal engaged or stop in the current lane (i.e., stop in its previous trajectory) to await an opportunity to change lanes and safely cross over.

[0206] IV.(w) ELV Convergence Liability An autonomous vehicle may behave differently depending on the regulations of the jurisdiction through which the autonomous vehicle is passing. In some cases, the laws may differ, including laws regarding the right of way between an autonomous vehicle and a manually operated or conventional vehicle in various situations. For example, under Arizona law, an autonomous vehicle has the right of way when an ELV merges. An autonomous vehicle may follow actions to avoid an accident while also complying with local laws and regulations.

[0207] IV.(x) Do not change lanes towards ELV

[0208] An automated vehicle must not change lanes into a lane adjacent to an ELV, including an ELV that was previously identified in the ELV memory and is now obscured.

[0209] IV.(y) Critical ELV Definition

[0210] A critical ELV can be defined as an ELV that poses a higher safety risk to both the autonomous vehicle and the ELV itself due to the condition, state, or location of the ELV.

[0211] The following types of ELVs can be considered critical ELVs:

[0212] ELV with (multiple) pedestrians nearby

[0213] ELV emergency vehicle

[0214] ELVs that protrude into the driving lane

[0215] Mobile ELV

[0216] ELV integrated into the return traffic

[0217] ELV with flashing lights (not hazard lights)

[0218] Large vehicles ELV (vehicle length >= predetermined distance (e.g. 8m, 9m, 10m)

[0219] All types of ELVs within a predetermined distance (e.g., 1.25m, 1.5m, 1.75m) from the driving lane.

[0220] IV.(z) Non-critical ELV Definition

[0221] Non-critical ELVs can be defined as ELVs that have a low probability of posing an emergency safety risk to both the autonomous vehicle and the ELV itself.

[0222] An ELV that is non-safety-critical and is more than a predetermined threshold distance (e.g., 1.0m, 1.25m, 1.5m, 1.75m) from the lane boundary but less than a predetermined upper limit distance (e.g., 3.2m, 3.4m, 3.6m, 3.8m, 4.0m) can be considered a non-critical ELV.

[0223] IV.(aa) Regulation ELV

[0224] Regulatory ELVs can be defined as ELVs that have legal issues that do not pose an imminent safety risk.

[0225] Any ELV that is not safety critical and is further away from the lane boundary than a predetermined threshold distance (eg, 3.4m, 3.6m, 3.8m, 4.0m) may be classified under this category.

[0226] IV.(ab) Initialization - Detecting Lane Changes

[0227] As a precaution, an autonomous vehicle traveling on a road with more than 2 lanes may perform a lane change to move away from the leftmost lane or the rightmost lane upon detecting the presence of an ELV (assuming the ELV position is unknown). The initial detection of an ELV where the ELV position is unknown may include any of the following: detection based on audio cues (e.g., siren detection); detection of patterns such as flashing lights; and detection of traffic movement ahead of the autonomous vehicle (e.g., traffic in the lane ahead is changing lanes or shifting in a particular direction), etc., where the actual ELV vehicle is not sensed by features, or an indication of the ELV is detected or sensed by the autonomous vehicle.

[0228] IV. (ac) Center Lane Preference

[0229] When the autonomous vehicle is not in the leftmost lane or the rightmost lane of the road, an autonomous vehicle traveling on a road with more than two lanes can avoid changing lanes to the outermost lane or the innermost lane as soon as the presence of the ELV is detected (assuming the position of the ELV is unknown).

[0230] IV.(ad)Wide Lane Merge Area Description

[0231] Fig.12An example of a wide lane merging zone is shown. A wide lane merging zone may be defined as when two adjacent lanes are parallel in the same direction, separated by a dashed line, and merge into a single lane greater than a predetermined width (e.g., greater than 4m in width). The start of the zone may be defined as the merging point, and the end may be defined as when the width of the merged single lane is less than a predetermined distance (such as 3.75m, 4m, 4.25m).

[0232] IV.(ae) Wide Lane Merge Area Exception When an ELV is present in a wide lane on a two-lane road merging into one lane and on the shoulder within a lane merging area, the automated vehicle may change lanes to the merging lane (i.e., the lane adjacent to the shoulder) and slow down and offset.

[0233] IV.(af)ELV before wide lane merge area

[0234] When the ELV is not within a wide lane and wide lane merging area (as described above), but within a predetermined number of meters (e.g., 75m, 100m, 125m) before the merging point, if conditions permit, the autonomous vehicle may only change lanes for critical ELVs (i.e., ELVs that pose a safety threat). For non-critical and regulatory ELVs (as described above) that are not within a wide lane or wide lane merging area, the autonomous vehicle may instead slow down and deviate. For example, when both the ELV and the autonomous vehicle are in a lane that does not end, the autonomous vehicle may only change lanes for critical ELVs. When both the ELV and the autonomous vehicle are in a lane that is about to end, the autonomous vehicle may change lanes in advance to avoid the ELV. In the case where the ELV is in a lane that does not end and the autonomous vehicle is in a lane that is about to end, the vehicle may change lanes to merge after passing the ELV. When the autonomous vehicle is in a lane that does not end and the ELV is in a lane that is about to end, the autonomous vehicle may remain in the same lane.

[0235] IV. (ag) ELV exception near planned exit routes

[0236] When approaching an exit on a planned route, the autonomous vehicle can remain in the exit lane for at least a predetermined number of meters (e.g., 700m, 800m, 900m) to avoid missing the exit. The autonomous vehicle can change lanes for critical ELVs, and slow down and offset for non-critical and regulatory ELVs in the zone.

[0237] If the exit is missed, the autonomous vehicle may MRC (i.e., perform a minimal risk maneuver such as exiting the lane, gradually stopping in its current lane, or suddenly stopping in its current lane), or the autonomous vehicle may choose an alternate route.

[0238] IV. (ah) Lane Change Decision Maker - Lane Deviation

[0239] If it is predicted that no portion of the autonomous vehicle will be able to laterally vacate the autonomous vehicle's half of the original lane on the ELV side by changing lanes before the time the autonomous vehicle closes the longitudinal gap with the ELV, the autonomous vehicle may decelerate and offset to the maximum extent (e.g., using a maximum offset that is extended beyond any lane line as necessary to avoid a collision) for all types of ELVs to maintain a lateral distance of at least a minimum number of meters (e.g., 2.5m, 2.74m, 3.0m, 3.25m) to the nearest lateral point of the ELV, rather than changing lanes. The autonomous vehicle may end the lane offset after completely passing the ELV. The autonomous vehicle may take action to prioritize avoiding collisions with the ELV or pedestrians around the ELV.

[0240] IV.(ai) Lane change in multi-lane merging area

[0241] Fig.13 An example scenario of a driving maneuver performed by an autonomous vehicle 1302 traveling alongside an ELV 1304 is shown. When the autonomous vehicle is traveling on a multi-lane road and is in a lane where an adjacent lane is about to merge into the lane of the autonomous vehicle, and the ELV is present on the shoulder of the road within a lane merging area, the autonomous vehicle may execute a lane change away from the ELV and into the adjacent lane if conditions permit and a lane change is available.

[0242] When an ELV is encountered in an area adjacent to the current lane being crossed, and the autonomous vehicle cannot change lanes to overtake according to predetermined criteria, then the autonomous vehicle may begin to decelerate and offset to maintain a safe speed and lateral distance no later than a predetermined number of meters (e.g., 75m, 90m, 110m, 125m, 152m, 160m, 170m) before reaching the ELV. The predetermined criteria that may cause the autonomous vehicle to begin decelerating and offsetting are as follows. The autonomous vehicle may be designed to change lanes as soon as the ELV is detected and begin to react no later than a predetermined number of meters (e.g., 125 meters (m), 130m, 140m, 150m, 152m, 155m, or 160m) before reaching the ELV.

[0243] In some scenarios, the autonomous vehicle may slow down and move away from the center of the lane it is currently traveling in (i.e., offset), or may simply offset within the lane. For example, the autonomous vehicle may slow down and may use offsets together, except for scenarios where only offsets are used for on-ramp scenarios and scenarios where the autonomous vehicle determines that the ELV is merging and intends to cut behind. In scenarios where the autonomous vehicle determines that the ELV is merging and intends to cut behind, the autonomous vehicle may maintain its current speed or reduce its speed minimally (e.g., slowing down by no more than 1 m / s 2) to indicate that it is passing the ELV. The automated vehicle may use the maximum passing speed in Table 5 below, or the minimum of (1) the value in Table 5 below and (2) a value 5 mph less than the maximum passing speed in Table 4 above to slow down for emergency lane pedestrians (people near or around ELVP-ELV), emergency lane emergency vehicles (ELEV), and ELVs with flashing lights.

[0244] Table 5

[0245]

[0246]

[0247] IV.(aj) ELV near speed limit change area

[0248] When encountering an ELV in a speed limit change zone, the autonomous vehicle may follow the following strategies depending on the state of the autonomous vehicle:

[0249] 1) Autonomous vehicles are accelerating

[0250] 2) Autonomous vehicles are slowing down

[0251] The new speed limit takes effect at speed limit signs.

[0252] IV. (ak) Behavior of autonomous vehicles accelerating near ELVs

[0253] Except when the autonomous vehicle is overtaking an ELV on a ramp segment, if the autonomous vehicle is accelerating due to an increased speed limit, it can stop accelerating when the ELV is within a predetermined number of meters (e.g., 75m, 90m, 110m, 125m, 152m, 160m, 170m) of the autonomous vehicle and overtake the ELV at a speed that does not exceed the maximum overtaking speed of the segment where the center of mass of the ELV is located.

[0254] When stopping acceleration, if the autonomous vehicle is faster than the maximum passing speed, the autonomous vehicle may decelerate, and if the autonomous vehicle is slower than the maximum passing speed, the autonomous vehicle may maintain a constant speed.

[0255] IV.(al) Acceleration stop zone

[0256] Fig.14An example acceleration stop zone is shown that may be adjacent to the location of a disabled or incapacitated vehicle. The autonomous vehicle may stop accelerating and maintain a constant speed when any portion of the autonomous vehicle is within the acceleration stop zone, except when the autonomous vehicle is on a ramp segment, in a traffic jam, or accelerating to change lanes. The acceleration stop zone is defined by a box extending longitudinally from the farthest point of the ELV to a predetermined distance (e.g., 30m, 35m, 40m, 45m, 50m) from the closest point of the ELV and laterally from the closest point of the ELV to a predetermined distance (3m, 3.25m, 3.5m, 3.75m, 4m) toward the vehicle of the autonomous vehicle.

[0257] IV. (am)ELV exception near single-lane on-ramps

[0258] If the ELV is present on a single-lane entrance ramp and the automated vehicle is accelerating to merge onto the highway, the automated vehicle may perform lane deviation and cease acceleration when it passes the ELV at a speed at which the automated vehicle can safely merge.

[0259] When deciding to exceed the speed of the ELV, the autonomous vehicle can consider the distance between the ELV and the merging point and the maximum possible acceleration of the autonomous vehicle within that distance (assuming the maximum acceleration is 0.35 m / s 2 、0.4m / s 2 , 0.5m / s 2 , 0.5m / s 2 any one of ).

[0260] IV. (an) accelerated recovery

[0261] The autonomous vehicle may resume acceleration after the rearmost point of the autonomous vehicle passes the frontmost point of the ELV.

[0262] IV. (ao)ELV exception near multi-lane on-ramps

[0263] Fig.15 An example of a driving-related maneuver performed by an automated vehicle operating on a multi-lane on-ramp of a highway is shown. When the automated vehicle encounters an ELV on the multi-lane on-ramp, the automated vehicle may avoid being in an adjacent lane of the ELV only if it does not prevent the automated vehicle from merging into highway traffic.

[0264] For example, when the ELV appears in the merging lane of an entrance ramp and the autonomous vehicle is in the merging lane, the autonomous vehicle can remain in the merging lane and offset with respect to the ELV. When the ELV appears in the merging lane of an entrance ramp and the autonomous vehicle is in the non-merging lane, the autonomous vehicle can remain in the same lane and change lanes to merge only after overtaking the ELV. When the ELV appears in the non-merging lane of the entrance ramp and the autonomous vehicle is accelerating to merge onto the highway, the autonomous vehicle can prioritize moving away from the ELV and changing lanes to the merging lane in order to safely overtake and merge. When deciding on the speed to overtake the ELV, the autonomous vehicle can consider the distance between the ELV and the merging point and the maximum possible acceleration of the autonomous vehicle within that distance (assuming a predetermined maximum acceleration, e.g., 0.3 m / s 2 、0.4m / s 2 , 0.5m / s 2 、0.6m / s 2 ).

[0265] IV. (ap) Behavior of autonomous vehicles slowing down near ELVs

[0266] When the automated vehicle is decelerating due to an anticipated speed limit reduction and a detected ELV, the automated vehicle may continue to decelerate and overtake the ELV at a speed intended not to exceed the maximum overtaking speed for the section of road where the ELV is located as specified in Table 4 (above).

[0267] IV.(aq) Deceleration Timing

[0268] Before the front bumper of the autonomous vehicle passes the closest point of the ELV longitudinally (relative to the autonomous vehicle), the autonomous vehicle may implement a targeted deceleration and offset strategy. The targeted deceleration and offset strategy is as follows. The autonomous vehicle may be designed to change lanes as soon as the ELV is detected and begin to react no later than a predetermined number of meters (e.g., 125 meters (m), 130m, 140m, 150m, 152m, 155m, or 160m) before reaching the ELV.

[0269] IV. (ar) Lane deviation level

[0270] When offsetting for an ELV, the autonomous vehicle may perform a critical safety offset for a critical ELV and a non-critical safety offset for non-critical and regulatory ELVs. A critical safety offset may be defined as follows. When an offset is available for a situation with an urgent safety issue, the autonomous vehicle may offset as much as possible away from the hazard while relaxing lane boundaries (if necessary), such that the maximum offset extends beyond any lane line as needed to avoid a collision. A non-critical safety offset may be defined as an offset as much as possible away from the hazard without relaxing lane boundaries.

[0271] IV.(as) Controlled speed deceleration

[0272] If the autonomous vehicle is traveling at a controlled speed that is less than the maximum passing speed for the road segment, the autonomous vehicle may decelerate by at least 5 mph from the controlled speed when performing a deceleration to indicate an intention to decelerate.

[0273] IV. (at) ELV approaches the driving lane

[0274] When encountering an ELV within a predetermined number of meters (laterally) of the autonomous vehicle (e.g., 1.0m, 1.3m, 1.5m, 1.75m), the autonomous vehicle may perform a full lane change or lane deviation to maintain at least a predetermined lateral distance (e.g., 1.0m, 1.2m, 1.3m, 1.4m, 1.5m) from the ELV. In this scenario, it is permitted for the autonomous vehicle to cross the following lane boundaries to perform a full lane change or avoid an accident.

[0275] For example, in Arizona, an AV may cross these lines to avoid an ELV: Dashed white line, Dashed yellow line, Solid white line. An AV may only cross the following lines to avoid a collision: Solid yellow line and double solid white line.

[0276] IV.(au) Stop and wait for ELV to close

[0277] If the ELV is predicted to be unable to maintain the minimum critical distance of 0.5 m from the ELV at the closest point of approach, the automated vehicle must not cross the ELV and may either change lanes completely or stop with the lane change signal engaged and wait for an opportunity to change lanes and cross safely.

[0278] IV. (av) NPC to ELV transition - behavior

[0279] If the vehicle is transitioning to an ELV and the autonomous vehicle is adjacent to the emergency lane, the autonomous vehicle may change lanes to move away from the current lane.

[0280] IV. (aw) NPC to ELV transition - lane deviation

[0281] If the autonomous vehicle is following a transition ELV in an adjacent lane on the shoulder and cannot change lanes, the autonomous vehicle may decelerate and maintain a preferred following distance from the transition ELV. The autonomous vehicle may use a deceleration and offset strategy to pass the transition ELV only after it has completely exited the autonomous vehicle's lane.

[0282] IV.(ax)ELV Confluence - Behavior

[0283] When the autonomous vehicle is driving in a lane adjacent to the ELV in which the ELV is moving, the autonomous vehicle may change lanes following the lane change priority of the lane change intention priority model so as to minimize interaction with the moving ELV (i.e., the vehicle transitions to become an NPC). The lane change intention priority model is as follows:

[0284] Lane change intentions are based on safety, regulatory and efficiency considerations unless otherwise specified. More specifically, the order of priority is as follows (from highest priority to lowest priority):

[0285] 1. Critical safety – Danger exists, i.e., an imminent threat to the safety of the autonomous vehicle

[0286] 2. Non-critical safety – danger exists

[0287] 3. Regulation – situations where legal bodies or regulations may govern the behavior of vehicles (including autonomous vehicles)

[0288] 4. Efficiency

[0289] 5. Prevention

[0290] 6. Preferences

[0291] If the automated vehicle cannot pass a complete lane between itself and the moving ELV, the automated vehicle may respond to the merging ELV based on a Cut Infront request and a Cut Behind request (described below), depending on whether the merging ELV is predicted to cut in front of or behind the automated vehicle, respectively.

[0292] IV.(ay)ELV merging - cutting in from the front

[0293] When a moving ELV is predicted to cut in front of the autonomous vehicle, the autonomous vehicle may accept the cut-in and maintain a preferred following distance with the NPC after the cut-in. A cut-in vehicle may be defined as a vehicle that partially or completely changes into the driving lane of the autonomous vehicle within a minimum gap distance. The minimum gap distance may be defined as a gap that ensures that the critical stopping distance (i.e., the distance between the rear of the front vehicle and the front end of the autonomous vehicle) is maintained when the vehicle in front of the autonomous vehicle (in this case, the cut-in ELV) brakes immediately and stops completely. The minimum gap may be based on the most conservative distance, which takes into account the reaction time of the autonomous system, the reaction time of the braking system, the maximum available deceleration of the system, the maximum possible deceleration characteristics of the leading vehicle based on the type (assuming the worst-case scenario of vehicle type, load, etc.), and the speed of the autonomous vehicle and the leading vehicle. The critical stopping distance may be defined as the distance required to avoid a collision between the autonomous vehicle and its surrounding vehicles when the autonomous vehicle stops suddenly.

[0294] IV.(az)ELV merging - cutting in from behind

[0295] When a moving ELV is predicted to cut in from behind, the automated vehicle may offset and overtake the ELV. Unless safety critical, the automated vehicle shall not decelerate beyond a threshold speed (e.g., 0.8 m / s) before and during signaling of the intent to overtake. 2 , 1m / s 2 , 1 5m / s 2 ). Offsetting and overtaking the ELV may include the automated vehicle initiating deceleration and offsetting to maintain a safe speed and lateral distance no later than a predetermined distance before reaching the ELV (e.g., such as 125 m, 140 m, or 152 m (500 ft) before reaching the ELV).

[0296] IV.(ba)ELV Memory

[0297] The autonomous vehicle may locate each ELV and store its position until the autonomous vehicle passes each corresponding ELV.

[0298] IV.(bb) Do not change lanes towards ELV

[0299] The automated driving vehicle may avoid changing lanes into lanes adjacent to an ELV, including an ELV that was previously identified in the ELV memory and is now obscured.

[0300] IV.(bc) Late detection of ELV

[0301] If the ELV is obscured from the field of view of the autonomous vehicle or is detected too late (e.g., high curvature areas, lane angle areas), and the system cannot meet the reaction distance requirements for safely changing lanes or decelerating and offsetting (e.g., starting deceleration and offsetting to maintain a safe speed and lateral distance no later than a predetermined distance before reaching the ELV (e.g., 125m, 140m, or 152m (500 feet) before reaching the ELV), the system can determine whether the autonomous vehicle can change lanes or perform lane offset based on the following criteria. If it is predicted that any part of the autonomous vehicle cannot laterally vacate half of the original lane of the autonomous vehicle on the side of the ELV by changing lanes before the time when the autonomous vehicle has closed the longitudinal gap with the ELV, then the autonomous vehicle can decelerate and offset to extend the maximum offset beyond any lane line as needed to avoid collision, thereby maintaining a predetermined lateral distance to the nearest lateral point of the ELV for all types of ELVs (e.g., at least 2m, at least 2.5m, at least 2.74m (9 feet)), instead of changing lanes. The autonomous vehicle can end the lane offset after completely passing the ELV.

[0302] IV. Continuous ELV in Group (bd)

[0303] The autonomous vehicle may treat multiple consecutive ELVs in the same lane within a predetermined number of meters (e.g., 20m, 25, 30m, 35m, 40m) between their closest points as a group and respond to them based on the most critical ELV in the group. The criticality of the ELV may be related to the danger posed to the autonomous vehicle by any given ELV.

[0304] IV. (be) straddling the lane

[0305] When the autonomous vehicle is on a two-lane highway and the ELVs on both sides of the road are within a longitudinal distance of a predetermined amount of curvature correction (e.g., 100m, 125m, 150m, 175m), and lane straddling is available, the autonomous vehicle may react no later than a predetermined number of meters (e.g., 100m, 125m, 152m, 175m, 200m) from the nearest ELV and perform lane straddling between the two lanes to maintain lateral equidistance between the ELVs.

[0306] IV.(bf) Lane straddling description

[0307] Lane straddling can be defined as when the autonomous vehicle is above the lane lines rather than between them and occupies more than one lane.

[0308] IV.(bg) Lane-crossing behavior

[0309] The autonomous vehicle may slow down and turn on hazard lights when lane straddling occurs. Once the autonomous vehicle passes the last ELV in the group, the autonomous vehicle may return to the original lane in which the autonomous vehicle was traveling before lane straddling began.

[0310] IV.(bh) Lane straddling availability

[0311] Lane straddling can be defined as available when the following conditions are met:

[0312] The autonomous vehicle is able to maintain a predetermined amount of bumper clearance (e.g., at least 10m, at least 15m, at least 17m, at least 20m) with the leading vehicles in both lanes that the autonomous vehicle intends to straddle.

[0313] The lane that the autonomous vehicle is intruding into does not have an object behind the autonomous vehicle with a bumper clearance of less than 12m and a collision time of less than 7s.

[0314] IV.(bi) Use of hazard lights

[0315] When the autonomous vehicle begins lane straddling, the autonomous vehicle may activate hazard lights, and may keep the hazard lights on until the autonomous vehicle fully returns to the autonomous vehicle's original driving lane.

[0316] Fig.19 An example flow chart of autonomous driving operations performed by a vehicle to operate on a road with an emergency vehicle is shown. Operation 1902 includes determining, by a computer located in the autonomous vehicle, that the emergency vehicle is within a predetermined distance of a first position of the autonomous vehicle operating in a lane on the road. Operation 1904 includes: in response to the determination, operating the autonomous vehicle to turn from the center of the lane toward a first side of the lane away from the center of the lane and away from the second position of the emergency vehicle, wherein the autonomous vehicle is caused to turn toward the first side until the lateral distance between the emergency vehicle and the autonomous vehicle is greater than or equal to the predetermined distance. In some embodiments, operating the autonomous vehicle to turn from the center of the lane as explained in operation 1904 includes sending instructions to one or more devices (e.g., one or more motors) in a steering system of the autonomous vehicle to cause the autonomous vehicle to turn.

[0317] In some embodiments, the method further comprises: in response to determining that the line separating the lane and a second lane immediately adjacent to the lane comprises a white dashed line, a yellow dashed line, or a white solid line, causing the autonomous vehicle to turn toward the first side of the lane and turn onto the second lane. In some embodiments, the method further comprises: in response to determining that the emergency vehicle is within a predetermined distance of the first position of the autonomous vehicle, and in response to determining that a lane change operation by the autonomous vehicle is not possible, sending an instruction that causes the autonomous vehicle to apply braking, or to decelerate the autonomous vehicle to a speed less than a threshold speed value.

[0318] In some embodiments, the threshold speed value is based on the rules of the region or state or region in which the autonomous vehicle is located. In some embodiments, the threshold speed value is based on the speed limit of the first location where the autonomous vehicle is operating and the lateral distance between the emergency vehicle and the autonomous vehicle. In some embodiments, the threshold is based on the speed limit of the first location where the autonomous vehicle is operating and the lateral distance between the emergency vehicle and the autonomous vehicle. In some embodiments, the autonomous vehicle operates to turn from the center of the lane toward the first side of the lane, and causes the autonomous vehicle to apply braking, or decelerate the autonomous vehicle to a speed less than the threshold speed value in response to the following circumstances: determining that the emergency vehicle is operating in another lane adjacent to the lane on which the autonomous vehicle is operating; and determining that a lane change operation of the autonomous vehicle is impossible. In some embodiments, the method also includes: operating the autonomous vehicle to accelerate only for lane change or to perform an avoidance maneuver in response to determining that the emergency vehicle is approaching the autonomous vehicle and the emergency vehicle is operating in another lane adjacent to the lane on which the autonomous vehicle is operating.

[0319] In some embodiments, the system further comprises a sensor subsystem, the sensor subsystem comprising a camera, a temperature sensor, an inertial sensor (IMU), a global positioning system, a light sensor, a LIDAR system, a radar system, and wireless communications, wherein the computer located in the autonomous vehicle is configured to utilize data from any sensor subsystem to perform determinations and operations. In some embodiments, the system further comprises a vehicle control subsystem in operative communication with the computer located in the autonomous vehicle, wherein the processor is configured to communicate with the vehicle control subsystem to perform the method, the method causing the autonomous vehicle to turn from the center of the lane toward the first side of the lane, and causing the autonomous vehicle to apply braking, or decelerate the autonomous vehicle to a speed less than a threshold speed value, in response to: determining that the emergency vehicle is operating in another lane immediately adjacent to the lane on which the autonomous vehicle is operating; and determining that a lane change operation of the autonomous vehicle is not possible. In some embodiments, the system also includes a vehicle control subsystem operably connected to a computer located in the autonomous vehicle, wherein the processor is configured to execute the method, the method also including: operating the autonomous vehicle via the vehicle control system to accelerate solely for the purpose of changing lanes or to perform an evasive maneuver in response to determining that an emergency vehicle is approaching the autonomous vehicle and that the emergency vehicle is operating in another lane that is immediately adjacent to the lane in which the autonomous vehicle is operating.

[0320] In some embodiments, the threshold speed value is based on: the rules of the region, state, or area in which the autonomous vehicle is located; and the speed limit of the first location where the autonomous vehicle is operating and the lateral distance between the emergency vehicle and the autonomous vehicle. In some embodiments, the method also includes: for the emergency vehicle transitioning to the emergency lane vehicle, the autonomous vehicle changing lanes to move away from a lane adjacent to the emergency lane; and slowing down and matching the speed of the emergency vehicle transitioning to the emergency lane vehicle by the autonomous vehicle until the emergency vehicle exits the current lane of travel of the autonomous vehicle. In some embodiments, the autonomous vehicle uses onboard sensors to detect any of the following to identify the emergency vehicle as a vehicle transitioning to the emergency lane: the use of a turn signal by the emergency vehicle to indicate a direction toward the shoulder of the road; a deviation or change in trajectory of the emergency vehicle; activation of flashing lights indicating an emergency vehicle, rescue vehicle, or law enforcement vehicle; a change in speed of the emergency vehicle; and direct communication from the emergency vehicle to the autonomous vehicle indicating the emergency vehicle's intent to move to the emergency lane or shoulder of the road.

[0321] V. Lane deviation

[0322] The autonomous truck is able to appropriately offset its position in the lane. Techniques for offsetting the autonomous vehicle may include moving the autonomous vehicle from the center of the lane to closer to the right edge or left edge of the lane in which the autonomous vehicle is traveling. The technical benefit of lane offset related maneuvers is that it may improve safety or comply with applicable regulations that may require the autonomous vehicle to move from the center of the lane.

[0323] The appropriate use and execution of lane departure related technologies by the autonomous vehicle can be performed by the compliance module (e.g. Figure 1 166 in the figure) and may include: identifying opportunities for lane excursions; defining lane excursions based on applicable regulations; identifying and adhering to maximum allowable excursions within a lane; identifying excursion times, including when to begin excursions and when to stop excursions; identifying excursion triggering conditions, such as the presence of vehicles primarily located in adjacent lanes or intrusive objects; and the like.

[0324] V.(a) When to end the offset

[0325] When offset by another vehicle, the autonomous vehicle may stop offsetting when the autonomous vehicle has passed a non-player character vehicle (NPC) or has been passed by an NPC and the bumper-to-bumper lateral distance is a predetermined value (e.g., 10 m).

[0326] When the autonomous vehicle is overtaking, the autonomous vehicle may measure from the NPC's front bumper to the autonomous vehicle's rear bumper.

[0327] When the autonomous vehicle is overtaking, the autonomous vehicle may measure from the autonomous vehicle's front bumper to the NPC's rear bumper.

[0328] V.(b) Non-critical safety deviation

[0329] When the offset is appropriate for a situation with a non-urgent safety issue, the automated vehicle may offset as much as possible away from the hazard without relaxing the lane boundaries as defined in the Maximum In-Lane Excursion description.

[0330] This requirement may apply to the following situations:

[0331] 1. When driving parallel to an Emergency Lane Vehicle (ELV) with no pedestrians (and we cannot change lanes or are in the process of changing lanes)

[0332] V.(c) Critical safety deviation

[0333] When deviation is appropriate for a situation with an urgent safety concern, the autonomous vehicle can maximally deviate away from the hazard while relaxing lane boundaries (if necessary) to avoid a collision.

[0334] This requirement may apply to the following situations:

[0335] 1. When driving parallel to a non-compliant vehicle (and the autonomous vehicle cannot change lanes or is in the process of changing lanes).

[0336] 2. When driving parallel to an oversized vehicle.

[0337] 3. When driving parallel to a large vehicle on a curve.

[0338] 4. When driving parallel to a large vehicle in a narrow lane.

[0339] 5. When driving parallel to an ELV with pedestrians (and the autonomous vehicle cannot change lanes or is in the process of changing lanes).

[0340] The preferred behavior in the above situation may be to perform a different maneuver (such as a lane change). The offset is only applied when the lane change (or other preferred behavior) cannot be performed, or when the automated vehicle is in the process of performing a behavior.

[0341] V.(d)Offset level

[0342] The autonomous vehicle may determine an excursion threshold based on a desired excursion level.

[0343] Standard offset: The standard offset can be defined as an offset that places the widest point of the autonomous vehicle at a predetermined number of meters lateral distance from the nearest edge of the lane boundary. This may only apply to lanes with a standard width (3.66m or 12 feet). For reference, in a standard width lane, a perfectly centered autonomous vehicle has a spacing of 0.51m (1 foot 8 inches) from the widest point of the autonomous vehicle to the nearest lane boundary. If the center of the lane is used as a reference point, this requirement indicates that the planned deviation from the center is less than the predetermined number of meters.

[0344] Maximum Lane Excursion Description: The maximum lane excursion can be defined as the maximum possible excursion that causes the widest point of the autonomous vehicle to be at the nearest edge of the lane boundary given the width of the current lane.

[0345] V.(e) Offset strategy when offset is not available

[0346] When offset is not available in any of the following situations, the automated vehicle may offset by half the distance required for the selected offset level without crossing the lane boundary and may leave that relative position by adjusting speed:

[0347] When driving parallel to a non-compliant semi-trailer truck.

[0348] When driving parallel to an oversized vehicle.

[0349] When offset from pedestrians.

[0350] In some embodiments, for other offset situations, the autonomous vehicle may not offset when offset is not available unless a collision is avoided. In some embodiments, for other offset situations, the autonomous vehicle may only offset when offset is available unless a collision is avoided.

[0351] Alternatively, the autonomous vehicle may deviate by more than half the amount required to avoid a collision. For example, when there are semi-trucks driving parallel to each side of the autonomous vehicle and one of the semi-trucks is out of compliance, the autonomous vehicle may deviate by half the maximum amount.

[0352] V.(e) Maximum deviation - relaxing lane boundary conditions

[0353] When an urgent safety issue exists, the autonomous vehicle may consider extending the maximum offset beyond any lane line as necessary to avoid a collision. The autonomous vehicle may minimize the extension of the maximum offset so that a collision is still avoided. The maximum offset must not put the autonomous vehicle at risk of striking the hard shoulder, striking a pedestrian, being involved in a responsible accident, running over significant road debris, or running over unknown objects. The autonomous vehicle may extend beyond the lane line only if it is not at risk of being involved in a responsible accident, striking the hard shoulder, striking a pedestrian, running over road debris, or running over other objects.

[0354] V.(f) Deviation when the vehicle is on both sides

[0355] The timeline (defined by when the offset starts and when the offset ends) that the autonomous vehicle can offset the vehicle is called an offset period. When there are parallel vehicles with overlapping offset periods on each side of the autonomous vehicle in the presence of a non-compliant turning vehicle, the autonomous vehicle can offset half the distance required for the selected offset level and can leave the relative position as quickly as possible by adjusting the speed. If the vehicles on both sides of the autonomous vehicle are turning non-compliantly or both vehicles are compliant, the autonomous vehicle can avoid offsetting and can leave the relative position by adjusting its speed.

[0356] V.(g) Efficiency offset

[0357] In an efficiency move, when an offset maneuver is available for situations where there is no major safety concern, the automated vehicle may offset as a precaution, the automated vehicle may offset by a standard amount, i.e., an amount that places the widest point of the automated vehicle a predetermined lateral distance (e.g., 0.2m, 0.25m, 0.28m (~11 inches), 0.3m) from the nearest edge of the lane boundary. This may only apply to lanes with a standard width (3.66m or 12 feet). This requirement may apply when driving parallel to a compliant semi-trailer truck on a non-curved lane of standard width (3.66m or 12 feet).

[0358] V.(h) Available offset

[0359] The autonomous vehicle may define an excursion as an available maneuver based on the level of excursion intent. The autonomous vehicle may preferably define an excursion as available when:

[0360] The automated vehicle is in the outside lane and there is no hard shoulder.

[0361] The automated vehicle is in the outside lane and no cars are merging onto the road at that point.

[0362] The autonomous vehicle is in the outside lane and there are no oncoming ELVs.

[0363] The automated vehicle is in the outside lane and there are no upcoming unknown objects or road debris on the shoulder.

[0364] The automated driving vehicle is in the middle lane and deviates beyond the lane boundary, meeting the lane change gap requirement in the planned deviation direction.

[0365] The autonomous vehicle is in the middle lane, offset within the lane boundaries, with parallel NPCs on both sides (limited offsets available).

[0366] The critical safety offset may meet the lane change gap requirement of the critical safety lane change intention.

[0367] V.(i) Description of the maximum deviation within the lane

[0368] The maximum in-lane excursion can be defined as the maximum possible excursion given the current lane width such that the widest point of the autonomous vehicle is at the nearest edge of the lane boundary. For a non-curved standard width lane (3.66m or 12 feet), the maximum in-lane excursion for a perfectly centered autonomous vehicle is 0.51m (1 foot 8 inches).

[0369] V.(j) Description of available offsets

[0370] The autonomous vehicle may define offsets as available in the following situations:

[0371] The automated vehicle is in the outside lane and there is no hard shoulder.

[0372] The automated vehicle is in the outside lane and no cars are merging onto the road at that point.

[0373] The autonomous vehicle is in the outside lane and there are no oncoming ELVs.

[0374] The automated driving vehicle is in the middle lane and deviates beyond the lane boundary, meeting the lane change gap requirement in the planned deviation direction.

[0375] The autonomous vehicle is in the middle lane, offset within the lane boundaries, with parallel NPCs on both sides (limited offsets available).

[0376] The critical safety offset can meet the lane change gap requirement of the critical safety lane change intention. A critical safety intention can be defined as an intention to change lanes due to an emergency safety issue. The following lane change intentions can be classified under this category: intention to change lanes at a predetermined distance (e.g., 75m, 100m, 125m, 150m) from the merging point of the merging area, predicting that the trajectory of the autonomous driving vehicle intersects with the trajectory of the merging vehicle in the merging area; intention to change lanes due to a moving ELV, ELVP (i.e., pedestrians around vehicles in the emergency lane), emergency vehicles, vehicles with flashing lights, abnormally stopped vehicles, or ELVs protruding into the lane of the autonomous driving vehicle; intention to change lanes to avoid a current acceleration <= -3m / s 2 Or expected future acceleration <= -2m / s 2 demanding trajectory; intention to change lanes to ensure that the autonomous vehicle does not miss the expected exit within a predetermined distance (e.g., 600m, 800m (~1 / 2 mile), 1000m, 1600m (~1 mile)); intention to change lanes because staying in the current lane traversed by the autonomous vehicle will result in an unexpected exit from the highway in less than a predetermined distance (e.g., 600m, 800m (~1 / 2 mile), 1000m, 1600m (~1 mile)); intention to change lanes due to height clearance constraints when the autonomous vehicle is too high, such an area should probably be classified as a no commercial vehicle zone.

[0377] V.(k) When does the deviation begin?

[0378] The autonomous vehicle may calculate when to begin offsetting based on the perceived position of surrounding vehicles. When offsetting is necessary due to another vehicle, the autonomous vehicle may begin offsetting when the bumper-to-bumper distance is (or less than) a predetermined lateral distance (e.g., 8m, 10m, 12m). When approaching a vehicle from behind, the autonomous vehicle may measure from the front bumper of the autonomous vehicle to the rear bumper of the NPC (i.e., the other vehicle).

[0379] When approached, the autonomous vehicle can measure from the NPC's front bumper to the autonomous vehicle's rear bumper.

[0380] V.(l) Audio notification when drifting

[0381] Whenever the autonomous vehicle plans to start an excursion, the system can make audio and human-machine interface (HMI) notifications, except for efficiency excursions (e.g., voluntary excursion maneuvers), where the autonomous vehicle can only make HMI notifications. In this patent document, an example of an HMI notification can include displaying a message on a display located in the autonomous vehicle.

[0382] The autonomous vehicle may only make 1 audio (if applicable) and 1 HMI notification per planned deviation instance. In addition, the notification HMI of the deviation maneuver may be repeated until confirmed by a human (such as a safety driver or test engineer).

[0383] V.(m) Standard deviation description

[0384] The standard offset may be defined as an offset that places the widest point of the autonomous vehicle at a lateral distance of 0.36m (1 foot 2 inches) from the nearest edge of the lane boundary. This may only be applicable to lanes with a standard width (3.66m or 12 feet).

[0385] For reference, a perfectly centered autonomous vehicle in a standard width lane would have a spacing of 0.51m (1 foot 8 inches) from the widest point of the autonomous vehicle to the nearest lane boundary.

[0386] V.(n) Offset Priority Model

[0387] Autonomous vehicles can prioritize and react to lane departure scenarios based on safety, regulatory, and efficiency considerations.

[0388] V.(o) Offset Timing

[0389] When an offset is necessary, the autonomous vehicle may begin offsetting before reaching the vehicle or object, and may end offsetting after the autonomous vehicle exceeds the vehicle or object by a safe distance.

[0390] V.(p) End excursion - cautiously return when vehicle is nearby

[0391] Figure 8 An example scenario is shown of the autonomous vehicle 802 returning to the center of a lane after performing a lane deviation maneuver when one or more vehicles 804, 806 are located in an additional lane adjacent to the lane in which the autonomous vehicle 802 is operating.

[0392] If the automated vehicle is returning to the center of the lane after a planned excursion and there are vehicles or other objects nearby in the planned lateral direction of movement, the automated vehicle may preferably converge to the center of the lane at a slower rate (e.g., using a trajectory indicated by a dashed line with an arrow) than if there were no vehicles or objects in the lateral direction of movement (e.g., using a trajectory indicated by a solid line with an arrow), unless a faster rate is required for obstacle / collision avoidance. A solid line with an arrow indicates a faster rate of returning to the center of the lane than a dashed line with an arrow, and a dashed line with an arrow indicates a more cautious (or slower) return to the center than a solid line with an arrow.

[0393] V.(q) Offset level prioritization - local road lane offset

[0394] The autonomous vehicle may define the offset by prioritizing available offset levels over desired offset levels.

[0395] V.(r) Available offset levels – local road lane offsets

[0396] The available offset level can be determined by the local road type on which the autonomous vehicle is located and the location of NPCs (e.g., other surrounding vehicles interacting with the autonomous vehicle) in that road segment.

[0397] One-way bicycle lane - inside the lane:

[0398] If the sidewalk or barrier is located on one side of the road, the automated vehicle may be permitted only to perform a maximum in-lane excursion (e.g., allowing an excursion maneuver to reach the lane line without crossing the lane line to avoid a collision) and to maintain at least a predetermined lateral distance between the outermost point of the automated vehicle and the outermost edge of the sidewalk or barrier.

[0399] Two-way road with reversible lanes:

[0400] The automated vehicle may be permitted to perform a maximum excursion - relaxing lane boundary conditions (e.g., allowing an excursion maneuver to extend beyond the lane line to avoid a collision) - up to half the lane width in an open reversible lane, provided that the reversible lane is free of NPCs within a predetermined number of seconds of time to collision (TTC) for the automated vehicle at the expected time that the automated vehicle fully exits the reversible lane.

[0401] Two-way road without a dividing strip:

[0402] The automated vehicle may be permitted to perform a maximum excursion - relaxing lane boundary conditions (e.g., allowing an excursion maneuver to extend beyond the lane line to avoid a collision) - up to a predetermined distance in the oncoming lane, provided that at the expected time that the automated vehicle fully exits the oncoming lane, there is no NPC in the oncoming lane within a TTC for a predetermined amount of time for the automated vehicle.

[0403] Multi-Lane:

[0404] An automated vehicle may be permitted to perform a maximum excursion—relaxing lane boundary conditions (e.g., allowing an excursion maneuver to extend beyond the lane line to avoid a collision)—up to half the lane width in an adjacent lane in the same direction of travel, provided that the lane is clear.

[0405] Bike Lanes:

[0406] The automated vehicle may be permitted to perform a maximum excursion - relaxing lane boundary conditions (e.g., allowing an excursion maneuver to extend beyond the lane line to avoid a collision) - up to a predetermined distance in a bicycle lane, provided that there are no NPCs within a curvature-corrected longitudinal distance of a predetermined number of meters (e.g., 100m, 125m, 150m, 175m) from the automated vehicle in the bicycle lane.

[0407] V.(s) Required offset level

[0408] The autonomous vehicle can define an offset distance based on the distance to the NPC and the conditions of the lane.

[0409] Overtaking scene:

[0410] When the autonomous vehicle is overtaking an NPC or is being overtaken by an NPC traveling in the same direction in an adjacent lane of a multi-lane road, the autonomous vehicle can perform a minimum non-critical safety offset of 0.23m from the center of the lane up to a maximum offset within the lane (e.g., allowing the offset maneuver to reach the lane line without crossing the lane line to avoid a collision).

[0411] The offset may be consistent with the standard offset used for non-critical safety offsets.

[0412] The autonomous vehicle can maintain a predetermined minimum lateral distance with an NPC traveling in the same direction in an adjacent lane of a multi-lane road.

[0413] Parking in adjacent driveways:

[0414] For a parked car in an adjacent lane, the autonomous vehicle may offset up to a maximum offset within the lane (e.g., allowing an offset maneuver to reach the lane line without crossing the lane line to avoid a collision), and the autonomous vehicle may maintain a predetermined minimum lateral distance from the parked car.

[0415] Parked car in adjacent lane - extended offset:

[0416] If the autonomous vehicle is in a narrow lane, or if it cannot maintain a predetermined minimum lateral distance from a parked car, the autonomous vehicle can perform a maximum excursion - relaxing the lane boundary conditions (e.g., allowing the excursion maneuver to extend beyond the lane line to avoid a collision).

[0417] Bicycles in the same lane:

[0418] If the autonomous vehicle determines that there is sufficient lateral distance to overtake a bicycle traveling in the same lane, the autonomous vehicle may comply with the maximum offset - relaxing the lane boundary conditions (e.g., allowing the offset maneuver to extend beyond the lane line to avoid a collision) to overtake the bicycle with a predetermined minimum lateral distance between the autonomous vehicle and the bicycle. The predetermined minimum lateral distance between the autonomous vehicle and the bicycle may be specified by local regulations or laws, or may be determined based on safety parameters of the autonomous vehicle. The predetermined minimum lateral distance between the autonomous vehicle and the bicycle may be a value in the range between 0.5m and 3m, such as between 1m and 2.5m, including between 1.25m and 2m.

[0419] Bicycles in bike lanes:

[0420] The autonomous vehicle may offset up to a maximum offset within the lane to overtake a bicycle moving in an adjacent bicycle lane, and the autonomous vehicle may maintain a predetermined minimum lateral distance from the bicycle.

[0421] Bicycle in bike lane - extended offset:

[0422] If the autonomous vehicle is in a narrow lane, or if it cannot maintain a predetermined minimum lateral distance from a bicycle, the autonomous vehicle can perform a maximum excursion - relaxing the lane boundary conditions.

[0423] Oncoming traffic:

[0424] When overtaking an oncoming vehicle in an adjacent opposing lane, the autonomous vehicle may offset up to a maximum offset within the lane, and the autonomous vehicle may maintain a predetermined minimum lateral distance from the oncoming vehicle.

[0425] Oncoming traffic – extended offset:

[0426] If the autonomous vehicle is in a narrow lane, or if it cannot maintain a predetermined minimum lateral distance to oncoming vehicles, the autonomous vehicle can perform a maximum excursion - relaxing the lane boundary conditions. The predetermined minimum lateral distance can be between 0.5m and 3.5m, between 0.75m and 3.0m, and between 1.0m and 2.5m.

[0427] Oncoming traffic offset timing:

[0428] The autonomous vehicle may start to deviate from the oncoming traffic NPC no later than a predetermined number of seconds of time to collision (TTC) before overtaking the oncoming traffic NPC. The predetermined number of seconds of TTC may be in the range of 1 to 12 seconds, 2 to 10 seconds, 5 to 9 seconds, such as 8 seconds.

[0429] Vehicles in reversible lane:

[0430] When overtaking a vehicle in a reversible lane, the autonomous vehicle may offset up to a maximum offset within the lane, and the autonomous vehicle may maintain a predetermined minimum lateral distance from the vehicle. The predetermined minimum lateral distance may be between 0.5m and 3.5m, between 0.75m and 3.0m, and between 1.0m and 2.5m.

[0431] Reversing Lane Vehicles - Extended Offset:

[0432] If the autonomous vehicle is in a narrow lane, or if a vehicle in the reversible lane intrudes into the lane of the autonomous vehicle, the autonomous vehicle may perform a maximum offset - relaxing the lane boundary conditions and overtaking the vehicle at a predetermined minimum lateral distance. The predetermined minimum lateral distance may be between 0.5m and 3.5m, between 0.75m and 3.0m, and between 1.0m and 2.5m.

[0433] Traffic at intersections:

[0434] If the automated vehicle is traveling straight in the intersection as part of through traffic, the automated vehicle may offset up to a maximum offset within the lane when avoiding a turning vehicle turning into an adjacent opposing lane, and the automated vehicle may maintain a predetermined minimum lateral distance from the turning vehicle and its predicted path. The predetermined minimum lateral distance may be between 0.5m and 3.5m, between 0.75m and 3.0m, and between 1.0m and 2.5m.

[0435] Intersection Traffic - Extended Offset:

[0436] If the autonomous vehicle is in a narrow lane, or if a turning vehicle is expected to intrude into the lane of the autonomous vehicle, the autonomous vehicle may perform a maximum excursion - relaxing the lane boundary conditions and leaving a predetermined minimum lateral distance between the paths of the autonomous vehicle and the turning vehicle. The predetermined minimum lateral distance may be between 0.5m and 3.5m, between 0.75m and 3.0m, and between 1.0m and 2.5m.

[0437] Stopped emergency vehicles:

[0438] When overtaking a stopped emergency vehicle, the autonomous vehicle may perform a maximum offset - relaxing lane boundary conditions, and the autonomous vehicle may maintain a predetermined minimum lateral distance from the emergency vehicle. The predetermined minimum lateral distance may be between 0.5m and 3.5m, between 0.75m and 3.0m, and between 1.0m and 2.5m.

[0439] Stopped Vehicles with Pedestrian or Hazard Lights:

[0440] When passing a stopped vehicle in an adjacent lane in a multi-lane road in the same direction of travel as a pedestrian or hazard light, the automated vehicle may perform a maximum excursion - relaxing the lane boundary conditions and the automated vehicle may maintain a predetermined minimum lateral distance from the vehicle or pedestrian, whichever is closer. The predetermined minimum lateral distance may be between 0.5m and 3.5m, between 0.75m and 3.0m, and between 1.0m and 2.5m.

[0441] ELV:

[0442] When the autonomous vehicle is unable to change lanes, the autonomous vehicle may offset up to a maximum offset within the lane (e.g., allowing an offset maneuver to reach the lane line without crossing the lane line to avoid a collision) to overtake the ELV, and the autonomous vehicle may maintain a predetermined minimum lateral distance from the ELV.

[0443] ELV – Extended offset:

[0444] If the automated vehicle is in a narrow lane, or if it cannot maintain a predetermined minimum lateral distance from a critical ELV, the automated vehicle may perform a maximum excursion—relaxing lane boundary conditions (e.g., allowing the excursion maneuver to extend beyond the lane line to avoid a collision).

[0445] Unknown Object:

[0446] When overtaking an unknown object in an adjacent lane, the autonomous vehicle may offset up to a maximum offset within the lane (e.g., allowing an offset maneuver to reach the lane line without crossing the lane line to avoid a collision), and the autonomous vehicle may maintain a predetermined minimum lateral distance from the unknown object.

[0447] Offset between NPCs:

[0448] When the autonomous vehicle is in a situation where there are NPCs on both sides of the autonomous vehicle that require offset, the autonomous vehicle can offset to ensure that the autonomous vehicle passes the two NPCs laterally at an equal distance.

[0449] Narrow Lane Description:

[0450] A narrow lane may be defined as any lane having a width less than a standard lane width or a predetermined distance, such as 3.5m, 3.45m, 3.4m, 3.35m or 3.3m.

[0451] V.(t) Deviation in heavy traffic

[0452] When the autonomous vehicle is in a traffic jam, the autonomous vehicle can avoid performing a maximum excursion—relaxing lane boundary conditions (e.g., allowing an excursion maneuver to extend beyond the lane line to avoid a collision).

[0453] V.(u) Map - Soft Border

[0454] For shoulders without guardrails, walls, or other physical barriers, the map can define a soft boundary as the boundary that separates any paved sections of the shoulder from any unpaved areas.

[0455] If there are bushes, trees, or branches that extend onto the paved road (up to the height of the autonomous vehicle), the map can define a soft boundary as the boundary that separates any paved section of the shoulder from the widest point of the bushes, trees, or branches.

[0456] V.(v) Maps – Hard Borders

[0457] For shoulders with guardrails, walls, or other physical barriers, the map can define the hard edge as the boundary that separates the paved section of the shoulder from the widest point of the physical barrier.

[0458] V.(w) Minimum lateral distance to the boundary

[0459] When offset within the lane boundary or offset beyond the lane boundary onto the shoulder or triangular road area, the widest point of the automated driving vehicle combination (e.g., including the tractor and trailer) can always maintain at least a predetermined distance (e.g., 30cm, 40cm, 50cm, 60cm) from the outermost point of any hard or soft boundary within a predetermined number of meters (e.g., 100m, 125m, 150m, 175m, 200m) longitudinally of the upcoming curvature correction relative to the foremost point of the automated driving vehicle combination.

[0460] V.(x) Loss of Global Navigation Satellite System (GNSS) – Lane deviation only:

[0461] In the event of interference, loss, or low confidence in the Global Navigation Satellite System (GNSS) or real-time dynamic positioning system signal, the autonomous driving vehicle can constrain the deviation to within the lane only to maintain the center of the lane, and may deviate from the center of the lane by no more than a predetermined distance (e.g., 0.2m, 0.3m, 0.4m).

[0462] V.(y) Loss of Global Navigation Satellite System (GNSS) - Return to original lane

[0463] In the event of interference, loss, or low confidence in GNSS or real-time dynamic positioning system signals, if the autonomous vehicle is currently deviating beyond the lane boundary, the autonomous vehicle may stop deviating beyond the lane boundary and return to its original lane unless doing so will result in a liability accident.

[0464] V.(z) Extension offset - Minimize time

[0465] The autonomous vehicle may drift beyond the lane boundary for the minimum amount of time required to avoid other vehicles or objects.

[0466] V.(aa) Extended offset - lateral distance to vehicles crossing the lane:

[0467] When traveling adjacent to another vehicle, the autonomous vehicle may extend beyond a lane boundary only if that vehicle crosses over the crossing lane boundary into the autonomous vehicle's current lane. In this regard, the autonomous vehicle may maintain a predetermined lateral distance (e.g., 1.0m, 1.15m, 1.3m, 1.5m) measured from the widest point of the autonomous vehicle combination to the widest point of the NPC. This requirement applies to lanes of all widths.

[0468] Extended offset - trajectory smoothing:

[0469] When the deviation exceeds the lane boundary of the neighboring vehicle, the autonomous vehicle can smooth its trajectory so that if the neighboring vehicle begins to swing back and forth, the autonomous vehicle will not imitate the swing.

[0470] Extended Offset - Return Lane:

[0471] When extending beyond the lane boundary of the neighboring vehicle, when the neighboring vehicle no longer intrudes into the original lane of the autonomous vehicle, or when the autonomous vehicle is no longer parallel to that vehicle, the autonomous vehicle may stop extending the offset and return to the original lane of the autonomous vehicle, as described below in “Return to Lane—No Fault Accident”.

[0472] Return to Lane - No Fault Accident:

[0473] When returning to its original lane after drifting beyond a lane boundary, the automated vehicle may seek to return to a longitudinal position that will not result in a liability-causing accident, adjusting speed if necessary to avoid a collision.

[0474] Extended offset - preferred offset when returning to the lane:

[0475] When returning to a lateral position within the lane boundaries, the automated vehicle may tend to deviate by a predetermined amount, such as specified in a lane deviation priority model.

[0476] V. (ab) Extension offset - preferred longitudinal behavior

[0477] When offset beyond the lane boundary of a vehicle in an adjacent lane, the autonomous vehicle may tend to the following behaviors.

[0478] When a vehicle in an adjacent lane is faster than the autonomous vehicle and the speed difference is greater than a threshold (e.g., 8mph, 10mph, 15mph), the autonomous vehicle can avoid accelerating to let that vehicle pass.

[0479] Conversely, when a vehicle in an adjacent lane is slower than the autonomous vehicle and the speed difference is greater than a threshold (e.g., 8mph, 10mph, 15mph), the autonomous vehicle may tend to overtake that vehicle.

[0480] Otherwise, the autonomous vehicle may tend to slow down unless it can overtake the vehicle in the adjacent lane within a predetermined number of meters (e.g., 100m, 125m, 150m, 175m, 200m) and the expected overtaking time is less than the expected deceleration time to exit parallel, in which case the autonomous vehicle does not need to slow down.

[0481] In these scenarios, overtaking means being able to overtake the vehicle to the extent that there is no overlap between the rearmost point of the automated vehicle combination (including the trailer) and the foremost point of the vehicle in the adjacent lane.

[0482] V. (ac) Extension deviation - a significant increase or decrease in the height of the road surface

[0483] The autonomous vehicle can ensure that it does not deviate beyond the lane boundary to the shoulder or lane angle area within a predetermined number of meters (e.g., 100m, 125m, 150m, 175m, 200m) of any (multiple) sections of the road where an increase or decrease in road surface height would cause control problems.

[0484] V.(ad) Extended offset - avoiding objects

[0485] The autonomous vehicle may only deviate beyond the lane boundary onto the shoulder or lane divider if there are no emergency lane vehicles (ELVs, i.e., vehicles in the emergency lane or shoulder), pedestrians, animals, road signs, or other objects parallel to or in front of the autonomous vehicle within a predetermined number of meters (e.g., 100m, 125m, 150m, 175m, 200m) or the distance required for a complete stop, whichever is greater. In a road environment, a lane divider is a space or area, usually triangular, defined by solid white lines of a through lane and an exit ramp or entrance ramp or road exit or entrance; a lane divider can help drivers organize when entering or leaving a highway and connect two areas of different elevations or slopes, such as the area between a ramp and a through portion of the road.

[0486] V.(ae) Extend offset - avoid object - return to original lane

[0487] When an ELV, pedestrian, animal, road sign, or other object appears within the distance required for the autonomous vehicle to come to a complete stop or within a predetermined number of meters (e.g., 100m, 125m, 150m, 175m, 200m), whichever is greater, and when the autonomous vehicle deviates beyond the lane boundary to the shoulder or lane divider, the autonomous vehicle may slow down, stop deviating beyond the lane boundary, and return to its original lane.

[0488] V.(af) Extended offset - lane split corners and merging vehicles

[0489] When the vehicle deviates beyond the lane boundary to the lane divergence corner area and there is a vehicle merging, the autonomous driving vehicle can slow down, stop deviating beyond the lane boundary, and return to its original lane before reaching the lane divergence corner point.

[0490] V.(ag) Extended offset - adjacent lanes

[0491] When offsetting beyond the lane boundary into an adjacent lane (local and highway) in the same direction of travel, the automated vehicle may be permitted to offset until the widest portion of the automated vehicle combination (automated vehicle with tractor and trailer) reaches a point where it is half the width of the adjacent lane, provided that that lane is clear.

[0492] V.(ah) Extended Excursion - Audio Notification

[0493] At the moment of known intent to deviate beyond the lane boundary, the autonomous vehicle may provide an audio notification "Extended Lane Departure" to notify the driver of its intent. The audio notification may be announced internally to the cab of the autonomous vehicle, or externally, such as to notify drivers of surrounding NPC vehicles of the autonomous vehicle's intent.

[0494] VI. Changing lanes

[0495] The self-driving truck is able to change lanes appropriately on a highway or road. Techniques for performing lane changes may include at least the following: identifying spaces (e.g., windows) in adjacent lanes that the self-driving truck can move into; monitoring vehicles in the lane that the self-driving truck wants to move into; identifying conditions where the lane change should be aborted; ensuring that lane changes and aborted lane changes are executed smoothly; and how to avoid collisions when changing lanes.

[0496] VI. (a) Order of Priority

[0497] When making decisions about lane changes, autonomous vehicles can prioritize safety over regulations, and both safety and regulations over efficiency.

[0498] VI. (b) Lane-changing intention priority model

[0499] The autonomous vehicle may classify lane change intentions based on safety, regulatory, and efficiency considerations, unless otherwise specified. More specifically, the priority order may be as follows (from highest priority to lowest priority):

[0500] Criticality Safety

[0501] A critical safety intention is an intention to change lanes possibly due to an emergency safety issue. The following lane change intentions may be classified under this category: an intention to change lanes at a predetermined distance from the merging point of the merging area (e.g., 75m, 100m, 125m, 150m from the merging point) where the trajectory of the automated vehicle is predicted to intersect with the trajectory of the merging vehicle in the merging area; an intention to change lanes due to a moving emergency lane vehicle (ELV), a pedestrian associated with an emergency lane vehicle (ELVP), an emergency vehicle, a flashing light vehicle, an abnormally stopped vehicle, or an ELV protruding into the lane of the automated vehicle; an intention to change lanes to avoid a demanding trajectory with a current acceleration less than or equal to a first value or a future expected acceleration less than or equal to a second value; an intention to change lanes to ensure that the automated vehicle does not miss its exit within a predetermined distance; an intention to change lanes because staying in our lane will result in an unintended exit from the highway in less than the predetermined distance; and an intention to change lanes due to a height clearance constraint in the case where the automated vehicle is too tall (this may be classified as a no commercial vehicle zone).

[0502] Non-critical safety

[0503] A non-critical safety intention is an intention to change lanes due to a safety issue that is unlikely to result in a collision.

[0504] The following lane change intentions may be classified under this category: an intention to change lanes between a first distance and a second distance from a merging point in a merging area, where the trajectory of the automated vehicle is predicted to intersect the trajectory of the merging vehicle in the merging area; an intention to change lanes due to the operation of an adjacent emergency lane vehicle (ELV) that is not safety critical; an intention to change lanes to avoid a demanding trajectory where the current acceleration is less than or equal to a first value or where the future expected acceleration is less than or equal to a second value; an intention to change lanes to ensure that the automated vehicle does not miss an expected exit between the first distance and the second distance from the automated vehicle; an intention to change lanes because remaining in the current lane of travel will result in an unintended exit from the highway within a range less than the first distance but greater than the second distance; and an intention to change lanes to bypass a slower vehicle traveling at a predetermined amount or more mph below the ambient speed.

[0505] Regulations

[0506] Regulatory intent is the intent that a lane change may be made due to legal issues that do not pose a major problem for safety.

[0507] The following lane change intentions may be classified under this category: an intention to change lanes because the automated vehicle is traveling in a lane that ends within a predetermined number of meters (e.g., end-of-lane merging); and an intention to change lanes because an area where commercial vehicles are not allowed (e.g., a no commercial vehicle zone) starts within a predetermined number of meters and is not a higher priority.

[0508] efficiency

[0509] Efficiency intention is the intention to change lanes for efficiency reasons.

[0510] The following lane change intentions may be classified under this category: an intention to change lane because staying in the current driving lane will cause the autonomous vehicle to unexpectedly exit the highway onto a local road within a predetermined number of meters (e.g., 1000m, 1100m, 1200m, 1300m, 1400m) from the highway onto a local road from which the autonomous vehicle can return to the highway via the local road; an intention to change lane because an upcoming planned exit is within a predetermined number of meters (e.g., 1200m (0.75 miles), 1600m (1 mile), 2000m (1.25 miles)) and a greater predetermined number of meters (e.g., 2800m (1.75 miles), 3200m (2 miles), 3600m (2.25 miles); intending to change lanes because staying in the current lane of travel will result in unintentional exit from the highway beyond a predetermined number of meters (e.g., 1200m (0.75 miles), 1600m (1 mile), 2000m (1.25 miles)) and a greater predetermined number of meters (e.g., 2800m (1.75 miles), 3200m (2 miles), 3600m (2.25 miles)); and intending to change lanes to bypass slow vehicles traveling between a first mph amount (e.g., 8mph, 10mph, 12mph, 15mph) and a second mph amount (e.g., 15mph, 18mph, 20mph, 25mph) below the ambient speed.

[0511] prevention

[0512] A preventive lane change intention is an intention to change lanes as a preventive measure.

[0513] The following lane change intentions may be classified under this category: an intention to change lanes between a first distance (e.g., 250m, 275m, 300m) and a second distance (e.g., 1150m, 1175m, 1200m, 1225m) from a merging point in a merging area, predicting that the trajectory of the autonomous driving vehicle intersects with the trajectory of the merging vehicle in the merging area; an intention to change lanes because the autonomous driving vehicle is traveling in a lane that ends (e.g., end-of-lane merging) between the first distance (e.g., 250m, 275m, 300m) and the second distance (e.g., 1150m, 1175m, 1200m, 1225m) from the autonomous driving vehicle; and an intention to change lanes to bypass a slow vehicle traveling within a predetermined value range below the ambient speed (e.g., between 5mph and 15mph, between 5.5mph and 12mph, between 6.7mph and 10mph below the ambient speed).

[0514] Preference

[0515] Preferential intention refers to the intention to change lanes because it does not pose a major problem for safety and is the lowest priority issue.

[0516] The following lane change intentions may be classified under this category: an intention to change lanes when there are more available right lanes than left lanes for the autonomous vehicle (e.g., right lane preference); and an intention to change lanes because a no-commercial-vehicle zone starts between a first distance and a second distance from the autonomous vehicle (e.g., between 50m and 1000m, between 75m and 900m, between 100m and 800m) and is not a higher priority.

[0517] VI. (c) Lane Change Rejection Priority Model

[0518] The autonomous vehicle may classify lane change rejections based on safety, regulatory, and efficiency considerations, unless otherwise specified. More specifically, the priority order may be as follows (from highest priority to lowest priority):

[0519] Criticality Safety

[0520] A critical safety denial is a denial where a lane change should not be performed due to an urgent safety issue.

[0521] The following lane change rejections may be classified under this category: lane changes where perception limitations would prevent the automated vehicle from safely changing lanes; lane changes that would place us into a lane with a height clearance constraint (which may be classified as a no commercial vehicle zone) if the automated vehicle is too high; lane changes that would move the automated vehicle out of a lane that the automated vehicle needs to enter in order to avoid an upcoming planned exit that is less than a predetermined distance away (e.g., 800m (0.5 miles), 1200m (0.75 miles), 1600m (1 mile)); lane changes that would move us out of a lane that we need to enter in order to avoid an unplanned exit that is less than a predetermined distance away (e.g., 800m (0.5 miles), 1200m (0.75 miles), 1600m (1 mile)). A lane change that moves the autonomous vehicle away from the incoming lane; a lane change that would move the autonomous vehicle away from a preconfigured trajectory for an upcoming intersection turn; a lane change that would place the autonomous vehicle in proximity to a moving emergency lane vehicle (ELV), an emergency lane vehicle near pedestrian (ELVP), an emergency vehicle, a vehicle with flashing lights, an abnormally stopped vehicle, or an ELV that protrudes into the autonomous vehicle's target lane; a lane change in which the autonomous vehicle does not have sufficient room to result in a high risk of collision; a lane change with a trajectory that is more demanding than the critical safety demanding trajectory intent that triggered it; and a lane change in which the target lane ahead vehicle is a slow-moving vehicle traveling at a predetermined amount of mph (e.g., 15mph, 18mph, 20mph, 25mph) or more below the ambient speed.

[0522] Non-critical safety

[0523] A non-critical safety rejection is a rejection that a lane change should not be made due to a safety issue that is unlikely to result in a collision.

[0524] The following lane change denials may be classified under this category: a lane change that would move the autonomous vehicle out of a lane that the autonomous vehicle would need to enter in order to avoid an upcoming planned exit that is greater than a first distance (e.g., 400 m (0.25 mi), 800 m (0.5 mi), 1200 m (0.75 mi)) but less than a second distance (e.g., 1200 m (0.75 mi), 1600 m (1 mile), 2000 m (1.75 mi), 2500 m (1.5 mi)); a lane change that would move the autonomous vehicle out of a lane that the autonomous vehicle would need to enter in order to avoid an upcoming planned exit that is greater than a first distance (e.g., 400 m (0.25 mi), 800 m (0.5 mi), 1200 m (0.75 mi)) but less than a second distance (e.g., 1200 m (0.75 mi), 1600 m (1 mile), 2000 m (1.75 mi), 2500 m (1.5 mi)). lane changes that would move the autonomous vehicle into a lane that is needed to exit an unplanned exit within 1600m (1 mile), 2000m (1.25 miles) of the road; lane changes that would place us adjacent to an ELV that is not safety critical; lane changes where the vehicle ahead in the target lane is a slow-moving vehicle traveling between a first mph amount (e.g., 5mph, 8mph, 10mph) and a second mph amount (e.g., 18mph, 20mph, 25mph) below the ambient speed; lane changes that would place the autonomous vehicle adjacent to a highway entrance ramp or an end-of-lane merging ramp that has a merging point that starts in less than a predetermined number of meters (e.g., 250m, 275m, 300m, 325m) and has at least one NPC on the ramp; and lane changes with a trajectory that is more demanding than the non-critical safety critical trajectory intention that triggered it.

[0525] Regulations

[0526] A regulatory refusal is a refusal where a lane change should not be made due to a legal issue that does not pose a major safety concern.

[0527] The following lane change rejections may be classified under this category: lane changes that would cause the autonomous vehicle to be in a no-commercial-vehicle zone that starts within a predetermined distance and is not a higher priority; lane changes that would cause the autonomous vehicle to cross a solid white lane boundary; and lane changes that would cause the autonomous vehicle to end in a lane within a predetermined distance (e.g., 1100m, 1150m, 1200m, 1250m).

[0528] efficiency

[0529] An efficiency rejection is a rejection where a lane change should not be made due to efficiency reasons.

[0530] The following lane change denials may be classified under this category: a lane change that would move the autonomous vehicle out of a lane that the autonomous vehicle would need to enter in order to avoid an upcoming planned exit that is between a first distance (e.g., 1200m (0.75 miles), 1600m (1 mile), 2000m (1.25 miles)) and a second distance (e.g., 2800m (1.75 miles), 3200m (2 miles), 3600m (2.25 miles)); a lane change that would move the autonomous vehicle out of a lane that the autonomous vehicle would need to enter in order to avoid an upcoming planned exit that is greater than the first distance (e.g., 1200m (0.75 miles), 1600m (1 mile), 2000m (1.25 miles)) and less than a second distance (e.g., lane changes that would result in an unplanned exit from the highway to a local road within a predetermined number of meters (e.g., 1200m (0.75 miles), 1600m (1 mile), 2000m (1.25 miles)), from which the automated vehicle can return to the highway.

[0531] prevention

[0532] A precautionary lane change rejection is a rejection where the lane change should not be made due to a relatively lower priority precaution.

[0533] The following lane change rejections can be classified under this category: lane changes that would place the autonomous vehicle adjacent to a highway entrance ramp or a lane-end merging ramp, where the merging point of the ramp starts within a first and second number of meters from the autonomous vehicle (e.g., 250m and 700m, 275m and 650m, 300m and 600m) and has at least one NPC on the ramp; lane changes that would cause the autonomous vehicle to drive parallel to a large vehicle (e.g., another semi-trailer truck).

[0534] Preference

[0535] A preference rejection is a rejection where a lane change should not be made because it does not pose a major concern to safety and is of the lowest priority.

[0536] The following lane change rejections can be classified under this category: lane changes that would result in the autonomous vehicle being in a no-commercial-vehicle zone that starts between a first number of meters (e.g., 75m, 100m, 125m) and a second number of meters (e.g., 700m, 800m, 900m) from the autonomous vehicle and that are not a higher priority.

[0537] VI.(d) Priority of conflict resolution between various modes

[0538] When faced with lane change intentions and lane change rejections of different priorities, the automated driving vehicle may follow the lane change intention or rejection with the higher priority, unless otherwise specified.

[0539] When faced with intents and rejections of equal priority, the autonomous vehicle can weigh the costs associated with all possible actions and choose the action with the lowest cost.

[0540] If it is possible to implement a lane change intention associated with a planned or unplanned exit while still being able to clear rejections of equal or lower level priority, the automated vehicle may be inclined to do so.

[0541] For example, when the automated vehicle is within a predetermined distance (e.g., 0.75 miles, 1 mile, 1.25 miles, 1.5 miles) of its exit and needs to move to the right lane. When there is a non-critical emergency lane vehicle (ELV) ahead, the automated vehicle may tend to pass the ELV before changing lanes to the right if it has room to do so while still being able to make the planned exit.

[0542] VI.(e) Continuously monitor the vehicle ahead of the target

[0543] The automated vehicle may continuously monitor the bumper-to-bumper clearance to the target front vehicle. The bumper-to-bumper clearance is defined as the distance in meters between the front bumper of the automated vehicle and the rear bumper of the target front vehicle. The target front vehicle is defined as the vehicle that is expected to be in front of the automated vehicle when the automated vehicle completes its lane change maneuver.

[0544] Slow down before predicting the expected target:

[0545] The autonomous vehicle is able to make an immediate prediction of the expected deceleration of the vehicle ahead in the target lane, if any. An immediate prediction is a prediction that uses any currently available information gathered from the sensors. For example, if our sensors detect heavy traffic ahead, the vehicle can expect to slow down by a certain amount.

[0546] Critical distance in front of target - Description:

[0547] The autonomous vehicle may define the critical distance to the target lane ahead vehicle as the maximum gap among: the bumper-to-bumper gap required to maintain high confidence in our sensor coverage; the bumper-to-bumper gap required outside our response time minimum; and the bumper-to-bumper gap required to avoid a collision under the assumption that both the autonomous vehicle and the target lane ahead vehicle must decelerate to a complete stop at the expected deceleration of the target lane ahead vehicle and the expected reaction deceleration of the autonomous vehicle. This gap may take into account the autonomous vehicle's reaction time and may include an additional safety buffer. This gap may be equal to the safety buffer when the target lane ahead vehicle is not expected to decelerate.

[0548] Critical distance in front of target - behavior:

[0549] An autonomous vehicle can avoid changing lanes within a critical distance from the vehicle ahead in the target lane.

[0550] Generally preferred behaviors ahead of the target:

[0551] For all lane change intentions except critical safety lane change intentions, the automated driving vehicle may prefer: a preference to change lanes with a bumper-to-bumper clearance of at least a predetermined distance (e.g., 10m, 12m, 15m) from the target ahead vehicle; a preference to follow the deceleration behavior outlined in "General deceleration behavior" in Section VI.(g); and a preference not to change lanes behind a slower-moving vehicle ahead of the target, as outlined in "Target lane slow vehicle behavior" in Section VI.(i).

[0552] Target front clearance growth:

[0553] For all lane change intentions, the autonomous vehicle may increase the front gap until the front gap is at an appropriate following distance as follows: after the outermost point of the autonomous vehicle's combined load has crossed the lane boundary of the target lane, the autonomous vehicle may maintain a positive front gap growth rate with the target front vehicle; and after the center of mass of the autonomous vehicle's combined load has crossed the lane boundary of the target lane, the autonomous vehicle may gradually increase the front gap growth rate with the target front vehicle to a predetermined rate (e.g., 1.5 m / s, 2 m / s, 2.25 m / s). Bumper-to-bumper gaps may be measured from the front bumper of the autonomous vehicle to the rear bumper of the target front vehicle.

[0554] VI.(f) Continuously monitor vehicles behind the target

[0555] The autonomous vehicle may continuously monitor the bumper-to-bumper clearance of the target rear vehicle. The bumper-to-bumper clearance is defined as the distance in meters between the rear bumper of the autonomous vehicle and the front bumper of the target rear vehicle. The target rear vehicle is defined as the vehicle that is expected to be behind the autonomous vehicle when the autonomous vehicle completes its lane change maneuver.

[0556] Determine the expected deceleration of the autonomous vehicle:

[0557] When considering a lane change, the autonomous vehicle is able to determine its expected deceleration if it were to change lanes

[0558] Critical distance behind target - Description:

[0559] The autonomous vehicle may preferably define the critical distance as the bumper-to-bumper clearance required to avoid a collision under the assumption that both the autonomous vehicle and the target lane rear vehicle must decelerate to a complete stop at the autonomous vehicle's expected deceleration and the target lane rear vehicle's expected reaction deceleration. This clearance may take into account the expected reaction time of the target rear and may include an additional safety buffer.

[0560] This gap should be equal to the safety buffer when the autonomous vehicle does not plan to slow down.

[0561] Critical distance behind target - behavior:

[0562] An autonomous vehicle can avoid changing lanes within a critical distance from the vehicle behind it in the target lane.

[0563] The following are the preferred behaviors for the target:

[0564] For all lane change intentions except critical safety lane change intentions, the autonomous driving vehicle may tend to change lanes in accordance with the listed conditions: when the bumper-to-bumper clearance with the target lane rear vehicle is at least a predetermined distance (e.g., 10m, 12m, 15m) and the collision time with the target rear is at least a predetermined number of seconds (e.g., 5s, 7s, 9s), the autonomous driving vehicle may tend to change lanes only when the speed of the target rear is greater than the speed of the autonomous driving vehicle; and when performing an efficiency, preventive, or priority lane change and the target rear vehicle is a large vehicle, the autonomous driving vehicle may tend to change lanes when the collision time with the target rear is at least a predetermined number of seconds (e.g., 8s, 10s, 12s).

[0565] General deceleration behavior behind the target:

[0566] For all lane change intentions except critical safety lane change intentions, the automated vehicle must not plan to slow down if the vehicle behind the target lane is outside our critical distance behind the target but within our preferred distance.

[0567] For all lane change intentions except critical safety lane change intentions, the automated driving vehicle may tend to change lanes in accordance with the listed conditions: the automated driving vehicle may tend to change lanes only when the speed of the target rear is greater than the speed of the automated driving vehicle, when the bumper-to-bumper clearance with the target lane rear vehicle is at least a predetermined distance (e.g., 8m, 10m, 12m, 15m) and the collision time with the target rear is at least a predetermined time (e.g., 5s, 6s, 7s, 8s); and when performing an efficiency, preventive, or priority lane change and the target rear vehicle is a large vehicle, the automated driving vehicle may tend to change lanes when the collision time with the target rear is at least a predetermined amount of time (e.g., 7s, 8s, 9s, 10s, 15s).

[0568] VI.(g) General deceleration behavior

[0569] For all lane change intentions except critical safety lane change intentions, when intending to change lanes, the automated driving vehicle may avoid planning to decelerate more than a predetermined rate (e.g., 2.24 m / s) when completing the lane change. 2 or approximately 3 to 6 mph per second, or approximately 4 to 5 mph per second). The expected deceleration of the autonomous vehicle can depend on the current speed and the expected deceleration of the object ahead.

[0570] VI.(h) Assisted Gap Finding

[0571] Under normal traffic conditions, autonomous vehicles can interact with nearby vehicles to assist in creating gaps for lane changes.

[0572] Gap finding turn signal usage:

[0573] The automated vehicle may activate appropriate turn signals when it intends to change lanes, unless the lane change is rejected by an equal or higher priority lane change rejection. When the automated vehicle is seeking a gap to change lanes, it may avoid immediately canceling the gap seeking intention due to insufficient space (critical safety lane change rejection), since the point of gap seeking is to signal and wait for space to be vacated.

[0574] Gap-finding turn signal duration:

[0575] For all lane change intentions, the autonomous vehicle may keep the signal illuminated until the autonomous vehicle successfully changes lanes or the gap seeking intention is canceled.

[0576] Identify vehicles that give way:

[0577] An autonomous vehicle can recognize when a vehicle in a target lane has yielded to it.

[0578] Yielding vehicles may flash their lights, create growing gaps, and / or maintain non-positive accelerations.

[0579] Cancel the intention - make the line gap too small:

[0580] For all lane change intentions except safety critical lane change intentions, if the vehicle is yielding, but (1) the rear gap is between our target rear critical distance and the preferred distance, and (2) the automated vehicle plans to slow down after the lane change, the automated vehicle may consider canceling the gap seeking intention.

[0581] Ultimately, the decision comes down to the cost of changing lanes within the autonomous vehicle’s preferred distance versus staying in the current lane of travel.

[0582] Cancel intent - Prefer gaps to find the duration of intent:

[0583] For efficiency-level lane change intentions, if a predetermined number of seconds (e.g., 15s, 18s, 20s, 22s) has passed since the turn signal was illuminated and the automated vehicle does not foresee a vehicle yielding, the automated vehicle may consider canceling the gap-seeking intention. For regulatory, non-critical safety, and critical safety lane change intentions, the amount of time that has passed since the turn signal was illuminated should not play a role in whether to cancel the gap-seeking intention.

[0584] Look for intent between retries:

[0585] If an efficiency-related gap-seeking intention is canceled due to duration (see “Canceling an Intent—Preferred Intent Duration”), the automated driving vehicle SHOULD consider retrying the intention after a predetermined number of seconds (e.g., 15s, 18s, 20s, 25s) has passed.

[0586] VI.(i) Continuous monitoring of slow-moving vehicles

[0587] An autonomous vehicle can continuously monitor the presence of slow-moving vehicles in its current lane and target lane.

[0588] Ambient speed:

[0589] The ambient speed may be the speed at which the autonomous vehicle would travel given its current environment, never exceeding the speed limit or contractual limit (if applicable).

[0590] The speed may take into account the speed limit, contractual limit (if applicable), traffic density, speed of most traffic and weather conditions. For free-flowing traffic in nominal weather conditions, the ambient speed may be the speed limit or contractual limit. "Free-flowing traffic" may include traffic with a level of service of A, B or C. Level of Service (LoS) is a term used to qualitatively describe road travel conditions based on factors such as speed, travel time, maneuverability, delays and safety. The level of service of a facility is represented by letters A to F, where A represents the best operating condition and F represents the worst operating condition.

[0591] Description of slow moving vehicles in the current lane:

[0592] If a vehicle in the current lane is traveling at a speed lower than the ambient speed by a predetermined amount of m / s or mph (e.g., 3 m / s (6.7 mph), 4 m / s (8.95 mph)) or more, the autonomous vehicle may consider that vehicle to be a slow vehicle.

[0593] Description of slow-moving vehicles in the target lane:

[0594] The autonomous vehicle may define a target lane slow-moving vehicle as any target ahead vehicle that would cause the autonomous vehicle to violate the general deceleration behavior requirements of its target behind or that is considered a slow-moving vehicle in the current lane after the autonomous vehicle has completed a lane change and is not expected to accelerate.

[0595] Predicted slowdown duration:

[0596] If there is a slow-moving vehicle in the autonomous vehicle's current lane or target lane, the autonomous vehicle can predict whether the vehicle will remain slow in the foreseeable future.

[0597] Some slow moving vehicles may be accelerating rapidly, which will affect whether they remain "slow" for a long time. When the speed limit is about to be reduced, marking slow vehicles can be based on the new slower speed limit.

[0598] Slow-moving vehicle observation window in the current lane:

[0599] When a vehicle in the autonomous vehicle's current lane is predicted to remain slow for a foreseeable amount of time (e.g., 10s, 12s, 15s, 18s), the autonomous vehicle may monitor the vehicle for at least a predetermined number of seconds (to allow it to accelerate) before considering overtaking.

[0600] Slow vehicle behavior in the current lane:

[0601] For slow-moving vehicles that are predicted to remain slow for an unforeseeable amount of time, the autonomous vehicle may consider overtaking.

[0602] If this type of slow moving vehicle is traveling less than a predetermined mph amount (e.g., 8mph, 10mph, 12mph) below the ambient speed, it should trigger an efficiency lane change attempt.

[0603] If this type of slow moving vehicle is traveling below the ambient speed by a predetermined amount of mph (e.g., 8 mph, 10 mph, 12 mph) or more, it may trigger a non-critical safety lane change intention.

[0604] Overtaking considerations:

[0605] When an autonomous vehicle is considering overtaking, it may tend to overtake on the left.

[0606] Passing a slower vehicle – Cooling down period:

[0607] When overtaking a slower vehicle, the autonomous vehicle may consider waiting until a predetermined number of seconds (e.g., 45s, 60s, 75s) before canceling the overtaking attempt (to allow the autonomous vehicle to complete the overtaking), unless a higher priority lane change attempt prompts the autonomous vehicle to another lane. In some embodiments, when the autonomous vehicle is overtaking a slower vehicle, the efficiency-related lane preference for the original lane should not take a higher priority (e.g., the autonomous vehicle should wait a predetermined number of seconds before giving up, slowing down, and returning to the original lane due to the efficiency-related lane preference).

[0608] Slow-speed vehicle behavior in the target lane:

[0609] If a slow-moving vehicle in the target lane is predicted to remain slow for an unforeseeable amount of time and travel less than a predetermined amount of mph below the ambient speed, the automated vehicle should classify it as an efficiency lane change rejection.

[0610] If the target lane slow-moving vehicle is predicted to remain slow for an unforeseeable amount of time and travel at a predetermined amount of mph or more below the ambient speed, the automated vehicle may classify it as a non-critical safety lane change rejection.

[0611] If possible, the autonomous vehicle may tend to overtake the slower vehicle in the target lane and then change lanes.

[0612] VI.(j) Lane change duration

[0613] The automated vehicle may complete all lane changes except critical safety lane changes within approximately a predetermined number of seconds. For critical safety lane changes or avoidance maneuvers, the automated vehicle may follow a minimum safe lane change duration defined by vehicle dynamics.

[0614] VI.(k) Avoiding collisions when changing lanes

[0615] An autonomous vehicle will never make a lane change that would result in a collision with another vehicle.

[0616] VI.(l) Monitoring the second lane of the vehicle

[0617] The autonomous vehicle may monitor the presence of vehicles adjacent to the autonomous vehicle's target lane (e.g., a vehicle two lanes away from the autonomous vehicle's current lane in the direction of the lane change) and may be inclined to change lanes if there are no vehicles in that location.

[0618] VI.(m) Abort Lane Change

[0619] Since the lane change takes more than ~8 seconds, the autonomous driving vehicle may prepare to abort the lane change if the original lane (e.g., the original driving lane) is still clear and prefers the original lane for safety reasons.

[0620] VI.(n) Stop Smoothness

[0621] When a lane change is aborted and there is no imminent collision risk, the autonomous vehicle may prefer to smoothly return to its original lane while changing lanes in approximately the same amount of time it took to reach its current position. If there is an imminent collision risk, the autonomous vehicle may prefer to follow a minimum safe abort duration defined by vehicle dynamics. As an example, if the autonomous vehicle has been changing lanes for 3 seconds when it decides to abort, it may take 3 seconds to return to its original lane.

[0622] VI.(o) Cooling Period Description

[0623] A cool-down period may be defined as a period of time after a lane change is completed in which another lane change cannot be initiated, typically to account for perceived limitations.

[0624] The lane change is considered complete when the automated vehicle is at the target lateral position in the target lane.

[0625] VI.(p) Cooling Period – Higher Priority Intent

[0626] If the current lane change intention has a higher priority than the previous lane change intention, the cooling period may last for a predetermined number of seconds, less than the number of seconds for the lower priority intention, and less than or equal to the number of seconds for the same level intention, unless otherwise specified. The predetermined number of seconds may be 2 seconds or less, including 1 second or less or 0.5 seconds or less.

[0627] VI.(q) Cooling Period – Lower Priority Intent

[0628] If the current lane change intention has a lower priority than the previous lane change intention, the cooling period may last for a predetermined number of seconds unless otherwise specified. The predetermined number of seconds may be 4 seconds or more, such as 5 seconds or more, 6 seconds or more, and 7 seconds or more.

[0629] VI.(r) Cooldown Period – Same Priority Intent

[0630] Except for critical lane change intentions and lane change intentions associated with slow vehicles, if the current lane change intention has the same priority as the previous lane change intention, the cooling period may last for a predetermined number of seconds, which may be 2s, 3s, 4s, or 5s.

[0631] VI.(s) Cooling period - Same priority intention - Critical lane change intention

[0632] If both the current and previous lane change intentions have critical safety priority, the cool-down period may last for a predetermined number of seconds equal to the number of seconds used for cooling down if the current lane intention had a higher priority than the previous lane.

[0633] For critical lane change intentions, the automated vehicle may want to ensure that it can react quickly enough, even if a previous intention had the same priority.

[0634] Other techniques for executing lane changes by the compliance module may include: continuous monitoring of slow moving vehicles; critical safety lane change intention actions; non-critical safety lane change rejection actions; non-critical safety lane change intention actions; actions for active lane change strategies when accepting merging vehicles on ramps: regulatory lane change rejection actions; efficiency lane change rejection actions; regulatory lane change intention actions; critical safety lane change rejection actions; and / or efficiency lane change intention actions

[0635] VII. Merge onto the road

[0636] Autonomous trucks can properly merge into lanes on the highway from on-ramps, or utilize k-ramps as appropriate. Performing merges on highways can include the ability to identify gaps in traffic that allow autonomous vehicles to merge, as well as the ability to identify when a vehicle should begin merging and complete the merge in accordance with applicable regulations.

[0637] VII.(a) Gap seeking - minimizing the change in planned acceleration

[0638] An autonomous vehicle may prefer to request the smallest gap for a planned change in longitudinal or lateral acceleration.

[0639] VII.(b) Gap Seeking – Narrowing Gap

[0640] An autonomous vehicle may tend to a gap whose size will not be closed. A gap with a target rear vehicle that intends to overtake the autonomous vehicle may be considered a closing gap even if there is no target front vehicle.

[0641] VII.(c) Merging speed preference

[0642] When merging onto a highway, the autonomous vehicle can accelerate to traffic speed before reaching the merging point, if possible.

[0643] VII.(d)k-Ramp-Alternative Route

[0644] The autonomous driving system may pre-map alternative routes for all k-ramps on the route of the autonomous vehicle that the autonomous vehicle may disengage from to merge. The alternative routes may outline paths to take in the event of an unsuccessful merge. Alternatively, the supervisory system may provide alternative routes for all k-ramps on the route that the autonomous vehicle may use to disengage from the road to merge.

[0645] VII.(e) Gap Creation

[0646] When merging onto a highway, the automated vehicle may proactively create a gap and flag vehicles for yielding, as described above with respect to lane change requirements.

[0647] An automated vehicle may not cancel an intent due to a narrow clearance gap and may therefore invalidate the requirement that, for all lane change intents other than critical safety lane change intents, the automated vehicle may tend to change lanes in accordance with the listed conditions: The automated vehicle should tend to change lanes when the bumper-to-bumper gap with the vehicle behind the target lane is at least a predetermined distance (e.g., 8m, 10m, 12m, 15m) and the collision time with the target behind is at least a predetermined time (e.g., 5s, 6s, 7s, 8s), which only applies when the speed of the target behind is greater than the speed of the automated vehicle.

[0648] When performing an efficiency, precautionary, or priority lane change, and the target rear vehicle is a large vehicle, Ego should tend to change lanes when the collision time with the target rear is at least a predetermined amount of time (e.g., 7s, 8s, 9s, 10s, 11s, 12s).

[0649] Merge gap creation in traffic jams:

[0650] When a merging gap cannot be found in a traffic jam, the autonomous vehicle can move forward slowly to find a potential merging gap.

[0651] When a potential merging gap is found, the autonomous vehicle may use its turn signal and move in the merging direction to a position where the autonomous vehicle is about to intrude into the final merging lane to indicate the intention to merge.

[0652] When the merging gap increases, the autonomous vehicle can continue to merge. When the merging gap decreases, the autonomous vehicle can stop and wait for vehicles to give way to the autonomous vehicle.

[0653] VII.(f) Gap Seeking – Maximizing Our Probability of Success

[0654] When merging onto the road, the autonomous vehicle can seek a gap that satisfies the critical distances in front of and behind the target (see the “Merge Gap” subsection) and maximizes the probability of a successful merge.

[0655] Gap Seeking - Minimizing the change in planned acceleration:

[0656] An autonomous vehicle may prefer to request the smallest gap for a planned change in longitudinal or lateral acceleration.

[0657] Gap Seeking – Closing Gap:

[0658] Autonomous vehicles may tend to fill gaps that won’t close in size.

[0659] Even if there is no target front vehicle, the gap to the target rear vehicle intending to overtake the autonomous driving vehicle can be considered as a narrowing gap.

[0660] Gap seeking – the trajectory of the leading vehicle:

[0661] When performing gap seeking, the autonomous vehicle may take into account the expected trajectory of any leading vehicle. In other words, the autonomous vehicle may not seek a gap that the leading vehicle is expected to enter and thus will not leave enough room for the autonomous vehicle.

[0662] VII.(g) Confluence and solid white line

[0663] When merging onto a highway, the automated vehicle can determine to avoid changing lanes when a tractor or trailer is parallel to a solid white line, unless it is doing so for an evasive maneuver.

[0664] Optionally, if necessary, when the tractor is not parallel to the solid white line but the trailer is parallel, the automated vehicle can initiate a merge onto the highway as long as the trailer stops being parallel to the solid white line before crossing the lane boundary.

[0665] Exception - Exiting the k-ramp and merging has already begun:

[0666] For a k-ramp, if the autonomous vehicle becomes parallel to the ending solid white line while in the process of changing lanes onto the highway (e.g., 1 / 2 of the tractor is on the highway and the other 1 / 2 is on the ramp), the autonomous vehicle may prefer to complete the lane change even if it needs to cross the solid white line.

[0667] VII.(h) Confluence time preference

[0668] An automated vehicle may prefer to merge onto the highway as quickly as possible while adhering to a merging trajectory that avoids or plans to require entering a lane split corner unless doing so is required for an evasive maneuver, and avoids arbitrary lane changes when a tractor or trailer is parallel to a solid white line unless doing so is required for an evasive maneuver.

[0669] VII.(i) Gap seeking - the trajectory of the leading vehicle

[0670] When performing gap seeking, the autonomous vehicle may take into account the expected trajectory of any leading vehicle. In other words, the autonomous vehicle may determine to avoid seeking a gap that the leading vehicle is expected to enter and thus will not leave enough room for the autonomous vehicle.

[0671] VII.(j) Merging and diverging angles

[0672] When merging onto a highway, the automated vehicle may determine to avoid or plan a merging trajectory that would require entering a lane split corner area unless doing so is for an evasive maneuver.

[0673] The tires of the autonomous vehicle touching the solid white boundary line can be regarded as "entering" the lane angle zone.

[0674] VII.(k) Confluence gap

[0675] When merging onto a highway, the autonomous vehicle may comply with the lane change requirements described herein, except that the autonomous driving system may determine to avoid having a preferred distance ahead of the target or a preferred distance behind the target. Conversely, when a merging gap is involved, the system may only comply with the critical distance requirements ahead of the target and behind the target.

[0676] VII.(l) k-Ramp Merge Priority

[0677] Exiting the k-ramp to merge onto the highway may be classified as a non-critical safety lane change attempt.

[0678] Given this classification, a lower priority rejection must not overrule the intent unless otherwise specified. A lower priority rejection may include a rejection for a scenario in which the autonomous vehicle monitors and detects the presence of a vehicle adjacent to the autonomous vehicle's target lane (i.e., a vehicle two lanes away from the current lane of travel in the direction of the lane change), and the autonomous vehicle may be inclined to change lanes when there is no NPC vehicle in that location.

[0679] VII.(m) Exception—Exiting the k-ramp and merging has already begun

[0680] For a k-ramp, if the autonomous vehicle becomes parallel to the ending solid white line while it is halfway through its lane change onto the highway (e.g., 1 / 2 of the trailer is on the highway and the other 1 / 2 is on the ramp), the autonomous vehicle can complete the lane change even if it needs to cross the solid white line.

[0681] The automated vehicle can still avoid crossing the lane angle in this situation unless it is for an evasive maneuver.

[0682] VII.(n)k-Ramp—Cancel Merge Attempt

[0683] When attempting to merge off a k-ramp, if the autonomous vehicle cannot find or create sufficient clearance when the front of the tractor has reached the solid white line or lane divider, the autonomous vehicle may cancel the merge attempt (and continue on the k-ramp).

[0684] VII.(o) Zipper Merge in Traffic Jam

[0685] When merging in a traffic jam, autonomous vehicles can follow zipper merging rules to achieve safety and public comfort.

[0686] VII.(p) Zipper Confluence Rules

[0687] The zipper merge rule can be defined as the behavior of vehicles in two merging lanes merging from each lane into the final lane in an alternating sequence of vehicles. This rule can be used for both low-speed and heavy traffic conditions when the merging gap is not large enough and the vehicles are slow enough to stop within a single car length.

[0688] VII.(q) Two lanes merge into one

[0689] If two lanes merge into one lane, the autonomous vehicle can execute a regulatory lane change from a predetermined distance from the end of the lane to the final merge into the lane, so that it can avoid being pushed to the end of the lane.

[0690] Lane Merge Sign Detection:

[0691] The autonomous vehicle is able to detect the lane merging sign from a predetermined distance and interpret which lane will be the final merging lane among the lanes.

[0692] Fig.21An example flow chart of an autonomous driving operation performed by a vehicle to merge onto a highway is shown. Operation 2102 includes: obtaining, by a computer located in the autonomous vehicle, an image from a camera located on the autonomous vehicle, wherein the image represents an area toward which the autonomous vehicle is driving on an entrance ramp of a highway. Operation 2104 includes determining, based on the image, that the area includes a merging section in which the autonomous vehicle on the highway is expected to merge into a lane on the highway. Operation 2106 includes: operating a turn signal to turn on in response to the determination, wherein the turn signal indicates that the autonomous vehicle is expected to merge from the entrance ramp into the lane on the highway. Operation 2108 includes: operating the autonomous vehicle to turn from the entrance ramp of the highway to the merging section on the lane of the highway in response to the determination and in response to the turn signal being turned on.

[0693] In some embodiments, the total length of the merging segment includes the length of the autonomous driving segment, the first minimum distance allowed between the autonomous driving vehicle and a first vehicle expected to be located in front of the autonomous driving vehicle, and the second minimum distance allowed between the autonomous driving vehicle and a second vehicle expected to be located behind the autonomous driving vehicle. In some embodiments, the method further includes: performing a first determination to determine that the length of the merging segment is decreasing; and in response to the first determination, operating the autonomous driving vehicle to apply brakes to stop the autonomous driving vehicle. In some embodiments, the method further includes: performing a second determination of a trajectory from an entrance ramp to the merging segment for the autonomous driving vehicle to follow in response to the determination, wherein the trajectory avoids the autonomous driving vehicle from entering a lane angle zone. In some embodiments, the image is acquired by the autonomous driving vehicle when it is determined from a priori images of other areas from a camera that the autonomous driving vehicle is driving toward that there is no additional merging segment and when the autonomous driving vehicle is operated to creep forward on the highway, wherein the priori images are acquired in time before the time when the image is acquired from the camera. In some embodiments, the autonomous driving vehicle is operated to creep forward at a speed less than a predetermined speed.

[0694] VIII. Parallel Large Vehicles and Objects

[0695] The autonomous truck may behave appropriately when approaching, being approached by, or driving parallel to a large vehicle. Such appropriate behavior encompasses the ability to determine that another vehicle is indeed driving or may soon be driving parallel to the autonomous vehicle, as well as the risks associated with situations where another vehicle is moving parallel to the autonomous vehicle. The appropriate behavior performed by the autonomous truck may be determined based on assessed risk, which may be affected by the duration of the parallel driving, the overall speed of the autonomous vehicle and surrounding traffic, and the desired route or trajectory of the autonomous vehicle. Appropriate behavior may include drifting in the current lane, changing lanes, slowing down, etc.

[0696] VIII.(a) Risk assessment for parallel driving with adjacent large vehicles

[0697] An autonomous vehicle can determine an acceptable amount of time to spend parallel driving with a neighboring large vehicle based on a risk model.

[0698] VIII.(b) Low risk description

[0699] An automated vehicle may define a low-risk parallel driving situation as one that is not a medium, high, or critical risk situation.

[0700] Lanes with normal width and where the offset is expected to be available for the duration of the parallel driving event may be considered low risk.

[0701] VIII.(c) Low-risk nominal behavior

[0702] In low-risk parallel driving situations, the autonomous vehicle may tend to be less than or equal to an expected parallel driving time that may be a predetermined time (e.g., 45s, 50s, 55s, 60s).

[0703] The autonomous vehicle may prefer the action that deviates the least from its planned action in the absence of a parallel truck. For example, the time it may take to overtake a neighboring truck with a speed difference of 0.89 m / s (2 mph) is about 55 seconds.

[0704] VIII.(d) Moderate Risk Nominal Behavior

[0705] In a medium risk parallel driving situation, the autonomous vehicle may tend to be less than or equal to an expected parallel driving time that may be a predetermined time (e.g., 30s, 36s, 40s).

[0706] For example, the time it would take to overtake a neighboring truck with a speed difference of, for example, 1.34 m / s (3 mph) would be approximately 36 seconds.

[0707] VIII.(e) Moderate Risk Description

[0708] An autonomous vehicle may define a medium risk parallel driving situation as one that satisfies any one or more of the following: the number of exits is reduced or will be reduced by more than 1; the lane is smaller than the regulatory width of the U.S. Interstate Highway System (3.6576m, 12 feet); it is expected that offset is not available for a portion of the parallel driving event; the road is curved; the situation may become higher risk if no action is taken; and situations where there are or will be large vehicles parallel on both sides of the autonomous vehicle can be considered medium risk.

[0709] VIII.(f) High Risk Description

[0710] An autonomous vehicle can define a high-risk parallel driving situation as a situation that meets any one or more of the following: the autonomous vehicle is currently parallel to a large vehicle that does not conform to the turning regulations or is within a predetermined distance (e.g., 12m, 15m, 18m) of being parallel to a large vehicle that does not conform to the turning regulations; and a situation that may become a critical risk if no action is taken.

[0711] VIII.(g) High-risk nominal behavior

[0712] In a high-risk parallel driving situation, the autonomous vehicle may tend not to enter the high-risk parallel driving zone unless the risk changes. If already in the zone, the autonomous vehicle may tend to transition out of the zone unless the risk changes.

[0713] Although the recommendations here are similar to critical risk situations, the autonomous vehicle may recognize that more urgent actions are required in critical risk scenarios, and high-risk scenarios may involve longer transition phases (such as longer deviations and decelerations before lane changes, etc.).

[0714] An autonomous vehicle can avoid high-risk parallel driving by any combination of the following actions: speeding up, slowing down, or changing lanes.

[0715] VIII.(h) Heavy Traffic Exception

[0716] If the traffic condition is at level of service D, E or F, i.e., poor to worst service levels as defined above, the automated driving vehicle may avoid following a low risk nominal behavior or a medium risk nominal behavior (as described above).

[0717] VIII.(i) Description of non-compliant truck

[0718] Generally speaking, non-compliant drivers and, by extension, non-compliant trucks are drivers, NPCs or trucks who do not comply with laws or regulations (such as right of way or speed limits), or who change lanes erratically, cut in sharply, use lane dividers or shoulders irregularly, etc.

[0719] VIII.(j) Parallel driving time prediction

[0720] The autonomous vehicle may predict the expected parallel driving time when a large vehicle in an adjacent lane approaches or is approached by a large vehicle in an adjacent lane.

[0721] VIII.(k) Description of approaching or being approached

[0722] The autonomous vehicle may define approaching or being approached as any time when our current planned action would cause the autonomous vehicle to parallel another large vehicle in an adjacent lane.

[0723] VIII.(l) Offsets for adjacent truck movements

[0724] When parallel to another large vehicle in an adjacent lane, the autonomous vehicle may offset, if available.

[0725] VIII.(m) Description of available offsets

[0726] The automated vehicle may define an offset as available when any of the following conditions are met: the automated vehicle is in the outside lane and there is no hard shoulder; the automated vehicle is in the outside lane and there are no vehicles merging onto the highway at that point; the automated vehicle is in the outside lane and there is no oncoming ELV; the automated vehicle is in the outside lane and there are no oncoming unknown objects or road debris on the shoulder; the automated vehicle is in the middle lane, offset beyond the lane boundary, and the lane change gap requirement is met in the direction of the planned offset; and the automated vehicle is in the middle lane, offset within the lane boundary, and there are parallel NPCs on both sides when limited offset is available.

[0727] VIII.(n) Description of parallel driving

[0728] The autonomous vehicle can define parallel driving as any time our system detects that the autonomous vehicle overlaps a large vehicle in an adjacent lane.

[0729] VIII.(o) Description of Large Vehicles

[0730] Regarding parallel large vehicles, if the length of the vehicle is greater than a predetermined length (e.g., 6m, 7m, 8m) or is an oversized vehicle, the autonomous vehicle may classify it as a large vehicle. The autonomous vehicle may avoid classifying livestock consumption vehicles (without attached trailers) as large vehicles.

[0731] For example, 2020 The F450 Double Cab - Long Wheelbase, can be considered a particularly long consumer vehicle at 6.8m. Due to these requirements, we would want to avoid classifying this consumer vehicle as a large vehicle unless it has a trailer attached.

[0732] VIII.(p) Critical Risk Nominal Behavior

[0733] In a critical risk parallel driving situation, the autonomous vehicle can exit the parallel driving through emergency and rapid actions. In some embodiments, faster maneuvers can take precedence over slower maneuvers.

[0734] VIII.(q) Improvement in the number of outlets

[0735] For low-risk and medium-risk scenarios only, the autonomous vehicle may avoid performing an intended action if performing the action would not result in the expected improvement in our number of exits.

[0736] In some embodiments, when an autonomous vehicle intends to change lanes, if the option is available, the autonomous vehicle may prefer to change lanes to an area that does not result in parallel driving with a large vehicle.

[0737] When an autonomous vehicle is driving parallel to a truck and intends to accelerate to pass, it may refrain from doing so if accelerating would cause the autonomous vehicle to drive parallel to another truck (unless the autonomous vehicle can pass both trucks within the times described in the low risk and medium risk requirements).

[0738] VIII.(r) Conflict mitigation risk actions

[0739] When faced with a decision involving two different parallel driving situations with different risk classifications, the autonomous vehicle can follow the nominal behavior of the higher risk classification.

[0740] VIII. Description of (s) "way out"

[0741] An autonomous vehicle can define “out” as any of eight zones around it: in front, behind, on both sides, and at the four corners.

[0742] The order of importance of the zones is as follows: rear > sides > front > four corners. That is, it is preferred to find an exit to the rear of the autonomous vehicle, then an exit to either side of the autonomous vehicle, then an exit to the front of the vehicle if either side is unavailable, then an exit to the four corners of the vehicle.

[0743] Having an empty rear zone is important because the self-driving vehicle can quickly slow down when needed. Having open sides is important for lane changes, however lane changes are usually riskier than slowing down, so these zones may not be as important as the rear zone. Having an open front is important, but acceleration for large trucks usually takes a lot of time, so its importance may not be as high as some other zones. Finally, corners are likely to be the least important because the self-driving vehicle cannot enter these zones directly without passing through another zone first.

[0744] VIII.(t) Criticality risk description

[0745] An autonomous vehicle can define a critical risk parallel driving situation as any situation where there is an imminent risk of safety liability.

[0746] Parallel driving when high winds could push the autonomous vehicle into a neighboring large vehicle can be classified as a critical risk.

[0747] Fig. 22 An example flow chart of an autonomous driving operation performed by a vehicle to operate adjacent to another vehicle is shown. Operation 2202 includes: acquiring, by a computer located in the autonomous vehicle, a set of images over time from a first camera located on the autonomous vehicle, wherein the set of images represents an area adjacent to a lane on which the autonomous vehicle is driving on a road. Operation 2204 includes: acquiring, by a computer, an image from a second camera located on the autonomous vehicle, wherein the image represents an additional area including the lane on which the autonomous vehicle is driving. Operation 2206 includes: performing a first determination based on the set of images to determine that the vehicle is driving adjacent to the autonomous vehicle for a length of time. Operation 2208 includes performing a second determination of a risk level associated with an autonomous vehicle driving parallel to the vehicle based on the image or set of images. Operation 2210 includes: performing a third determination in response to the first determination and the second determination to determine that the length of time is greater than a predetermined time period. Operation 2212 includes operating the autonomous vehicle to accelerate or decelerate in response to the third determination. In some embodiments, operating the autonomous vehicle to accelerate or decelerate includes sending an instruction to an engine to accelerate or decelerate, or sending an instruction to an actuator in a brake unit to apply a brake.

[0748] In some embodiments, performing the second determination includes: determining that the risk level is low in response to determining from the image that the width of the lane is within a range of predefined standard widths for a standard lane, and in response to determining that a trajectory for the autonomous vehicle to turn away from the center of the lane to one side of the lane is available, wherein the predetermined time period is associated with the low risk level. In some embodiments, performing the second determination includes: determining that the risk level is medium in response to determining from the image that the width of the lane is less than a range of predefined standard widths for a standard lane, or determining that a trajectory for the autonomous vehicle to turn away from the center of the lane to one side of the lane is not available, or determining that the lane includes a curved portion; wherein the predetermined time period is associated with the medium risk level.

[0749] In some embodiments, performing the second determination includes: determining that the risk level is high in response to determining that the autonomous vehicle is parallel to or within a certain distance of the turning vehicle based on the image set; and wherein the method further includes operating the autonomous vehicle to accelerate or decelerate or change lanes in response to the determination. In some embodiments, the method further includes: determining that the length of the vehicle is greater than a predetermined length based on at least one image in the image set; and in response to the determination, operating the autonomous vehicle to turn away from the center of the lane to one side of the lane.

[0750] IX. Pedestrian and / or Bicyclist Interactions

[0751] Self-driving trucks can identify, classify, and interact appropriately with pedestrians and cyclists. Each jurisdiction (e.g., state, country) may have its own regulations that any vehicle must comply with when operating around pedestrians and / or cyclists. Some regulations are high-level, such as avoiding intrusions into crosswalks or bike lanes. Other regulations are more detailed and depend on the relative location of the pedestrian or cyclist and the trajectory of the vehicle. For example, when a vehicle is turning from one road to another and there is a dedicated lane for such a turn, regulations may dictate how to interact with cyclists in a bike lane or path adjacent to the turning lane. In order for the self-driving truck to operate appropriately in accordance with applicable regulations, the compliance module of the self-driving truck (e.g., Figure 1 166 in the figure) can determine which regulation(s) apply based on the location and interaction type. In some embodiments, the compliance module can not only determine the location of the autonomous vehicle (e.g., based on the location provided by a GPS device on the autonomous vehicle), but can also identify pedestrians and / or cyclists, and can track the movement of pedestrians / cyclists relative to roads and lanes or dedicated surrounding areas (e.g., crosswalks, trails, bike lanes).

[0752] IX.(a) Minimum lateral distance when passing pedestrians / cyclists

[0753] When passing a pedestrian / cyclist, the autonomous vehicle may maintain a minimum lateral distance of at least a predetermined distance (e.g., 0.91 m or 3 feet) from the widest point of the autonomous vehicle (e.g., on the side of the autonomous vehicle facing away from the pedestrian) to the widest point of the pedestrian (e.g., on the side of the pedestrian facing away from the autonomous vehicle) until the entire tractor and trailer has passed the pedestrian / cyclist.

[0754] IX.(b) Do not cross a crosswalk where there are pedestrians

[0755] The front bumper of an automated vehicle must not cross a crosswalk where pedestrians or cyclists are crossing.

[0756] For example, crosswalks in Arizona or Texas span both directions of traffic, and in these states, autonomous vehicles may not cross the crosswalk when pedestrians or cyclists are located on the half of the road in which the autonomous vehicle is traveling.

[0757] In another example, for a crosswalk that spans two directions of traffic in California or New Mexico, an automated vehicle may not cross the crosswalk when pedestrians or cyclists are on either half of the roadway.

[0758] IX.(c) Yielding to Bicyclists in the Right Turn Lane

[0759] like Fig.10As shown, the autonomous vehicle 1002 may yield to the cyclist 1004 when approaching a right turn lane / reducing lane.

[0760] IX.(d) Following distance for pedestrians / pedestrians

[0761] If the autonomous vehicle cannot overtake a pedestrian or cyclist while traveling on local roads, where the distance, speed, or other safety constraints between the autonomous vehicle and the pedestrian or cyclist are as described herein, the autonomous vehicle may maintain at least a predetermined amount of following distance from the leading pedestrian or cyclist and match the speed of the pedestrian or cyclist.

[0762] If changing lanes to avoid a pedestrian is not possible because the pedestrian may be a law enforcement officer, the autonomous vehicle may stop due to a pedestrian on the highway.

[0763] IX.(e) Continuous Perception Distance of Pedestrians

[0764] On highways and local roads, the compliance module is able to consistently and accurately detect pedestrians or cyclists at distances greater than or equal to the autonomous vehicle’s current stopping distance.

[0765] Although this perception distance represents a lower limit of detection, there may be other maneuvers that require a greater perception distance. Therefore, the automated vehicle can adjust its continuous perception distance for pedestrians to meet the increased perception distance required for other maneuvers.

[0766] IX.(f) Visibility Control

[0767] When necessary, the system can adjust its speed so that it can come to a complete stop within the current visibility.

[0768] Visibility distance is the maximum distance at which the driver of a vehicle, including an autonomous driving system, can see and identify objects around the vehicle.

[0769] IX.(g) Use of Speakers

[0770] The autonomous vehicle may determine that it is best to avoid using the horn when driving near a cyclist. A sudden loud blast from the horn may startle the cyclist and cause them to swerve into traffic. Driving near a cyclist may be considered to be within 2m, 3m, 3.5m, or another suitable distance, which may be changed depending on the volume of the autonomous vehicle's horn.

[0771] IX.(h) Cyclist hand signals

[0772] Fig.11An identification of gestures determined by an autonomous vehicle and corresponding meanings is shown so that the autonomous vehicle can respond to cyclist gesture signals. The autonomous driving system of the autonomous vehicle can use camera and other sensor data in combination with a computing module to determine the presence of a cyclist to identify the gesture signals made by the cyclist. The knowledge provided by a map or mapping module on the autonomous vehicle can help identify the gesture signals by perhaps increasing the likelihood that the cyclist will use gesture signals to change direction or slow down. Alternatively or additionally, the autonomous vehicle can send data to a remote operator at a supervisory or control center for confirming the cyclist using gesture signals, and the remote operator can send information back to the autonomous vehicle via a communication module at the supervisory system and the autonomous vehicle. The information sent from the remote control operator or supervisory system can include the type of signal, the type of action that the autonomous vehicle should take, any changes in the trajectory of the autonomous vehicle, etc.

[0773] IX.(i) Do not overtake vehicles waiting for pedestrians to pass

[0774] At marked or unmarked crosswalks, automated vehicles must not pass / overtake stopped vehicles that are waiting for pedestrians to cross.

[0775] IX.(j) Pedestrian Memory

[0776] The compliance module of an autonomous vehicle may record in memory the presence of a pedestrian or cyclist that is later fully or partially obscured from view.

[0777] IX.(k) Yield to pedestrians

[0778] At roundabouts, intersections, and marked or unmarked crosswalks, automated vehicles may yield to pedestrians and cyclists who intend to cross into the path of the automated vehicle.

[0779] IX.(l) Preferred lateral distance when passing pedestrians / cyclists

[0780] When overtaking a pedestrian / cyclist on a highway or local road, the autonomous vehicle may tend to overtake with a lateral distance of at least a predetermined number of meters (e.g., 3.66m or 1 standard lane width) from the widest point of the autonomous vehicle to the widest point of the pedestrian until the tractor and trailer have completely overtaken the pedestrian / cyclist.

[0781] The behavior of autonomous vehicles when passing pedestrians / cyclists on highways or local roads can be guided by recommendations developed by transportation departments (e.g., the Arizona Department of Transportation (ADOT)).

[0782] When a pedestrian or cyclist is detected on the shoulder or corner of a highway, the compliance module of the autonomous vehicle may prefer that the autonomous vehicle travel in a lane that is not adjacent to the shoulder / corner.

[0783] IX.(m) Pedestrian deceleration and deflection strategies

[0784] When a pedestrian or cyclist is detected on a highway or local road, if the autonomous vehicle cannot meet the preferred lateral distance requirement (e.g., the distance from the autonomous vehicle to the pedestrian / cyclist), it can slow down and offset away from the pedestrian / cyclist. For example, if the compliance module determines based on sensor data (e.g., images) from a sensor (e.g., a camera) that a pedestrian or cyclist is located on the highway or local road, then the compliance module can send instructions to a device on the autonomous vehicle's equipment (e.g., a braking system, a steering system, etc.) to slow down and offset the autonomous vehicle (or cause the autonomous vehicle to move (e.g., turn) from approximately the center of the lane on the highway or road) away from the pedestrian or cyclist.

[0785] The maximum amount that the autonomous vehicle can offset while relaxing lane boundaries to move away from pedestrians / cyclists.

[0786] An autonomous vehicle can adjust its speed based on proximity to pedestrians / cyclists as outlined below:

[0787] Table 6

[0788]

[0789] When there are emergency lane vehicles (ELVs) in addition to pedestrians, the recommended maximum overtaking speed for the automated vehicle may be the minimum of (1) the values ​​in the above table discussed with respect to emergency lane vehicles on a highway or road and (2) a value that is 5 mph less than the corresponding value in Table 1 or Table 2 above.

[0790] IX.(n) Pedestrian control signals

[0791] The automated vehicle may yield the right of way to a pedestrian crossing in a marked or unmarked crosswalk using a pedestrian control signal that indicates or symbolizes that the pedestrian may “go.” The automated vehicle may use its sensors and detection modules or mapping and position information, or a combination of sensors and detection modules and mapping and position information, to determine that a pedestrian may be crossing in a marked or unmarked crosswalk using a pedestrian control signal that indicates or symbolizes that the pedestrian may continue to cross.

[0792] IX.(o) Avoid collision with pedestrians

[0793] Automated vehicles must not collide with pedestrians or cyclists, even when the pedestrian or cyclist does not have the right of way.

[0794] IX.(p) Stopping distance prediction

[0795] An autonomous vehicle is able to predict the distance it will need to come to a complete stop given its current driving condition and environmental conditions.

[0796] IX. (q) Do not follow pedestrians on the road

[0797] The autonomous vehicle may avoid planning trajectories that would cause the autonomous vehicle to drive directly behind a pedestrian or cyclist on a highway.

[0798] IX.(r) Partial occlusion

[0799] Autonomous vehicles are able to detect partially occluded pedestrians. Partial occlusion detection for autonomous vehicles can be benchmarked against partial occlusion detection for humans.

[0800] Fig.16 An example flow chart of an autonomous driving operation performed by a vehicle operating on a road or highway including pedestrians and / or cyclists is shown. Operation 1602 includes: acquiring, by a computer located in an autonomous vehicle, an image from a camera located on the autonomous vehicle, wherein the image represents an area toward which the autonomous vehicle is driving on a lane on the road or highway. Operation 1604 includes determining, based on the image, that a pedestrian or cyclist is located next to a lane on the road or highway. Operation 1606 includes: in response to the determination, operating the autonomous vehicle to turn from the center of the lane to a first side of the lane away from the center of the lane and away from the location of the pedestrian or cyclist. In some embodiments, the autonomous vehicle is operated to turn by sending instructions to one or more devices (e.g., one or more motors) in a steering system of the autonomous vehicle to turn the autonomous vehicle. Operation 1608 includes: in response to the determination, operating the autonomous vehicle to reduce the speed of the autonomous vehicle to below the first threshold speed value in response to determining that the lateral distance from the autonomous vehicle to the pedestrian or cyclist is within a first distance set and the current speed of the autonomous vehicle is greater than the first threshold speed value. In some embodiments, the autonomous vehicle is operated to reduce speed by sending instructions to one or more devices (e.g., one or more actuators of an engine or a braking unit) to apply brakes or decelerate the autonomous vehicle.

[0801] In some embodiments, the autonomous vehicle is caused to reduce the speed of the autonomous vehicle by comparing the lateral distance from the autonomous vehicle to the pedestrian or cyclist and the current speed of the autonomous vehicle to a table including a plurality of distance sets and a plurality of threshold speed values, wherein the plurality of distance sets include a first distance set and a second distance set greater than or equal to the first distance set, wherein the plurality of threshold speed values ​​include a first threshold speed value and a second threshold speed value greater than the first threshold, and wherein the first distance set and the second distance set correspond to the first threshold speed value and the second threshold speed value, respectively. In some embodiments, the first threshold speed value is a minimum of a first predetermined speed value and a first speed value, the first speed value being obtained by subtracting a specific speed from a speed limit, and the second threshold speed value is a minimum of the second predetermined speed value and the first speed value.

[0802] In some embodiments, the method further includes: operating the autonomous vehicle to maintain the speed of the autonomous vehicle in response to determining that the lateral distance from the autonomous vehicle to the pedestrian or cyclist is greater than a third distance set, the third distance set being greater than or equal to the second distance set. In some embodiments, the method further includes, in response to determining the presence of an emergency vehicle on the road or highway based on the image: operating the autonomous vehicle to reduce the speed of the autonomous vehicle to below a third threshold speed value in response to determining that the lateral distance from the autonomous vehicle to the pedestrian or cyclist is within the first distance set and the current speed of the autonomous vehicle is greater than a third threshold speed value, wherein the third threshold speed value is the minimum of the first threshold speed value and the maximum passing speed value.

[0803] In some embodiments, the maximum overtaking speed value is a specific speed that is less than the speed value, and wherein the speed value is based at least on a speed limit of a road or highway and whether the autonomous vehicle is operating on the road or highway. In some embodiments, the method further comprises: operating the autonomous vehicle to overtake a pedestrian or cyclist by maintaining a minimum lateral distance between the autonomous vehicle and the pedestrian or cyclist, wherein the minimum lateral distance is a predetermined distance from a side of the autonomous vehicle farthest from the pedestrian or cyclist to the position of the pedestrian or cyclist. In some embodiments, the pedestrian or cyclist is determined based on an image when a first distance from a first position of the autonomous vehicle to a second position of the pedestrian or cyclist is greater than or equal to a stopping distance of the autonomous vehicle, and wherein the stopping distance is a second distance required for the autonomous vehicle to come to a complete stop.

[0804] X. Turn signal

[0805] Autonomous trucks may use turn signals appropriately to safely traverse routes. Appropriate use of turn signals may require identifying any applicable regulations and taking action in accordance with those regulations. Appropriate use of turn signals may also include identifying an impending turn in the autonomous vehicle's trajectory and activating and terminating the signal in a manner that effectively warns surrounding drivers and vehicles. More complex maneuvers or combinations of maneuvers may require more detailed sub-features or tasks to perform or implement those features. Appropriate use of turn signals may also include identifying when a turn signal is not to be used or when a turn signal is terminated. For example, a turn signal may be terminated when a lane change or turn is no longer desired. Alternatively or additionally, early initiation of a turn signal may be prohibited when the autonomous truck wants to prevent a following vehicle from entering a gap into which the truck intends to merge.

[0806] X.(a) Signal Status Integrity

[0807] Turn signals may be used only to indicate turning intention.

[0808] When an automated vehicle is performing a lane reduction in the through lane (e.g., a 2-to-1 merge), the automated vehicle must not use a turn signal.

[0809] An automated vehicle must not use turn signals as a tool to prohibit passing traffic if the automated vehicle does not intend to turn.

[0810] X.(b) Use of pre-drawn turn signals - speed reduction

[0811] If the turn signals are not already illuminated, the autonomous vehicle can engage the appropriate turn signals as soon as it begins to decelerate due to an upcoming pre-mapped turning maneuver (e.g., if the compliance module determines that the autonomous vehicle is decelerating when the autonomous vehicle's position is within a predetermined distance of another location where the autonomous vehicle will perform a turning maneuver).

[0812] This requirement may not apply when the deceleration is a direct result of factors other than the upcoming maneuver, such as heavy traffic.

[0813] X.(c) Fault signal

[0814] If the compliance module on an autonomous vehicle detects that the turn signal is not working, the autonomous vehicle must not be driven until it is repaired, unless the autonomous vehicle is already on the road, in which case the autonomous vehicle may refer to a specified MRC (Minimum Risk Condition) maneuver.

[0815] X.(d) Chain steering - same direction

[0816] For a turn signal timeline that includes turn signal illumination for a turning maneuver as a turn signal illumination period, when two or more back-to-back turning maneuvers have overlapping turn signal illumination periods and the maneuvers are in the same direction (e.g., a right lane change followed by a right turn), the automated driving vehicle may continue to keep the appropriate turn signal illuminated until the last maneuver is completed.

[0817] X.(e) Description of Steering Control

[0818] A steering maneuver can be defined as a lane change, highway merge, highway exit, intersection turn, or any other planned action that will result in the automated vehicle driving in a lane different from its current lane.

[0819] X.(f) Turn signal when the maneuver is canceled

[0820] When a steering maneuver is canceled, the automated vehicle may turn off the turn signal unless the canceled maneuver is immediately retried (e.g., attempted again).

[0821] X.(g) Use of undrawn turn signals

[0822] For turning maneuvers that are not pre-mapped, the automated vehicle may turn on the appropriate turn signal once the intent of the maneuver is known, unless otherwise specified. Once the maneuver is completed, the automated vehicle may turn off the turn signal, unless another turning maneuver causes the automated vehicle to keep the signal illuminated.

[0823] Under Arizona law, New Mexico law, and Texas law, an autonomous vehicle may signal continuously for a predetermined number of meters (e.g., at least 30.5 m (100 feet)) before turning.

[0824] Examples of steering maneuvers that are not pre-mapped include lane changes for slower moving vehicles, emergency lane vehicles (ELVs), and avoiding adverse merging interactions.

[0825] X.(h) Use of pre-drawn turn signals

[0826] For steering maneuvers that the autonomous vehicle's compliance module knows in advance (such as due to knowledge of a map) that it will have to perform, the autonomous vehicle can proactively illuminate the appropriate turn signal before reaching the point where the maneuver is likely to begin. The autonomous vehicle can then turn off the turn signal once the maneuver is completed, unless another steering maneuver causes the autonomous vehicle to keep the signal illuminated.

[0827] X.(i) Chain steering - different directions

[0828] In a scenario where the turn signal is illuminated for a turning maneuver as the turn signal illumination period, when two or more back-to-back turning maneuvers have overlapping turn signal illumination periods, and the maneuvers are in opposite directions (e.g., a right turn at an intersection followed by a left lane change), the automated driving vehicle may fully complete the first maneuver before changing the signal for the second maneuver.

[0829] X.(j) Complete the steering operation

[0830] When the autonomous vehicle is at the planned destination of the steering maneuver, the autonomous driving system of the autonomous vehicle can mark the steering maneuver as completed.

[0831] X.(k) Turning beyond the intersection

[0832] When an autonomous vehicle plans to make a turning maneuver outside of an intersection, it can wait until its rear bumper passes the middle of the intersection before engaging the turn signal. In this case, engaging the signal too early may cause other drivers to enter the path of the autonomous vehicle.

[0833] X.(l) Completing the Turning Maneuver - Turning at an Intersection with a Slippery Lane

[0834] When making a right turn at an intersection with a sliding lane, the automated vehicle may mark the maneuver as complete when the tractor and trailer pass the apex of the sliding lane.

[0835] X.(m) Pre-drawn turn signal - lane change

[0836] When the autonomous vehicle is approaching a pre-mapped lane change, the autonomous vehicle may activate the turn signal at a predetermined distance or time to collision (TTC) from the starting point of the lane change (e.g., 150 m (~500 ft) or 5 s when traveling at 75 mph).

[0837] X.(n) Pre-drawn turn signal - intersection turn

[0838] When the autonomous vehicle is approaching a pre-mapped intersection and turning, the autonomous vehicle may turn on the turn signal when it is a predetermined distance (e.g., 60 m (~200 ft)) from the stop line of the intersection.

[0839] X.(o) Complete steering maneuvers - lane changes and highway merging

[0840] When turning for a lane change or highway merge, the automated vehicle can mark the maneuver as complete when the tractor and trailer are fully in the target lane.

[0841] X.(p) Pre-drawn turn signal - lane splitting

[0842] When the autonomous vehicle is approaching a pre-mapped lane split and intends to enter a ramp on a lane adjacent to the lane split corner, the autonomous vehicle can turn on the turn signal when it is a predetermined distance (e.g., 150m (~500 feet)) from the lane split corner.

[0843] X.(q) Completing the Turning Maneuver - Intersection Turning without Sliding Lanes

[0844] When (1) turning left at an intersection or (2) turning right at an intersection without a sliding lane, the automated vehicle may mark the maneuver as complete when the tractor and trailer are no longer in the intersection and are fully in the target lane.

[0845] X.(r) Completing Turning Maneuvers—Highway Exits, Local Exits, and Lane Diversions

[0846] When turning for a highway exit, local exit, or lane split, the automated vehicle may mark the maneuver as complete when the rear end of the trailer passes the lane split corner of the exit or lane split.

[0847] X.(s) Pre-drawn turn signal lights – highway exits

[0848] When the autonomous vehicle is approaching a pre-mapped exit from the highway, the autonomous vehicle may activate the turn signal when it is a predetermined distance from the lane corner, e.g., 200 m (~650 ft or 6 s at 75 mph).

[0849] X.(t) Pre-drawn turn signal - highway merging

[0850] When the autonomous vehicle is approaching a pre-mapped merge on a highway and is in a lane adjacent to a lane split corner, the autonomous vehicle may activate a turn signal when it is a predetermined distance from the lane split corner (e.g., 300 m (~1000 ft or 10 s when traveling at 75 mph).

[0851] When the merging ramp is less than a predetermined distance (e.g., 200m, 250m, 300m) and the autonomous vehicle is in the situation just described, the autonomous vehicle may activate the appropriate turn signal only when the tractor and trailer are completely on the ramp.

[0852] Fig.17 An example flow chart of an autonomous driving operation performed by a vehicle to operate a turn signal is shown. Operation 1702: Determine, by a computer located in the autonomous vehicle, that the autonomous vehicle is decelerating when the autonomous vehicle is at a first position on a road, the first position being within a predetermined distance of a second position at which the autonomous vehicle is to perform a turning maneuver. Operation 1704 includes: in response to the determination and in response to determining that the turn signal is not on, operating the turn signal to turn on at a first time. In some embodiments, operating the turn signal includes sending an instruction to turn on the turn signal.

[0853] In some embodiments, the method further includes sending an instruction in response to determining that the turn signal is not working or operating, the instruction causing the autonomous vehicle to turn to the side of the road along the trajectory and apply brakes. In some embodiments, the turn signal is caused to be turned on for a first time length at a first time, and wherein the method further includes: performing a first determination to determine that the first time length overlaps with a second time length, the second time length being associated with a second steering maneuver occurring after the steering maneuver; performing a second determination to determine that the second steering maneuver is in the same direction as the steering maneuver; and in response to the first determination and the second determination, operating the turn signal to remain on during the first time length and the second time length. In some embodiments, the predetermined distance is based on laws or regulations of the region or state in which the autonomous vehicle is operating.

[0854] In some embodiments, the method also includes: performing a determination to determine that the second location where the autonomous vehicle is to perform a steering maneuver is adjacent to, within a certain distance of, or passing through the intersection; and wherein in response to the determination, in response to determining that the turn signal is not turned on, and in response to determining that the rear of the autonomous vehicle passes through the middle of the intersection, causing the turn signal to be turned on.

[0855] XI Stop vehicle

[0856] XI.(a) Key Terms

[0857] Stopped Vehicle: A vehicle that can be stationary or immobile for any duration. A compliance module on an autonomous vehicle can determine whether a vehicle is stationary or immobile based on image processing performed on images acquired by a camera located on the autonomous vehicle.

[0858] Abnormally stopped vehicle: A stopped vehicle that may be stopped for reasons unrelated to traffic congestion or regulatory signs / signals. Alternatively or additionally, the stopped vehicle may be moving but is not moving, and is stopped for reasons unrelated to traffic congestion or regulatory signs / signals. For example, a stopped vehicle at a traffic light that has turned red is not an abnormally stopped vehicle, while an emergency vehicle that has stopped in the middle of a highway due to an emergency situation is an abnormally stopped vehicle. The compliance module on the autonomous vehicle can determine whether a vehicle (e.g., an NPC vehicle) is an abnormally stopped vehicle by performing image processing on the image(s) acquired by the camera and determining that there is no indication of traffic congestion in the image or that there are no traffic signals or traffic signs in the image(s).

[0859] Protruding abnormal stopped vehicle: An abnormal vehicle that intersects more than one lane (including the shoulder); this type of stopped vehicle protrudes into at least one other lane. An abnormal stopped vehicle that intersects more than one lane (including the shoulder). For example, this type of stopped vehicle protrudes into at least one other lane.

[0860] Abnormally stopped vehicle boundary area: An area that contains all abnormally stopped vehicles that are in close proximity to each other. If there is only one abnormally stopped vehicle, the boundary area is equivalent to the space occupied by that vehicle. Vehicles can be within a predetermined number of meters longitudinally (e.g., 20m, 30m, 40m) and a certain distance laterally (e.g., 7.0m, 7.3m, 7.5m) of each other to be considered in the same boundary area.

[0861] Traffic Jam: A column of road traffic at or near a standstill. A traffic jam may be further defined using a threshold speed for all lanes of traffic and / or a threshold maximum distance between each successive vehicle.

[0862] To be considered a traffic jam, the average speed of vehicles in all lanes within a predetermined number of meters (e.g., 125m, 150m, 175m, 200m) of the autonomous vehicle may be below a threshold speed (e.g., 8mph, 10mph, 12mph).

[0863] To be considered a traffic jam, a visible lane may contain a column of vehicles where the average bumper-to-bumper gap between each consecutive vehicle is less than a predetermined number of meters (e.g., 8m, 9m, 10m, 12m, 15m).

[0864] XI.(b) Exceeding the minimum lateral distance for stopped vehicles (STY)

[0865] The autonomous vehicle may decelerate and pass the abnormally stopped vehicle boundary zone only if the autonomous vehicle can maintain a lateral distance of at least a predetermined threshold within the boundary zone.

[0866] The predetermined threshold distance may be an adjustable parameter, and a nominal value thereof is any one of 1.0 m, 1.2 m, 1.3 m, 1.4 m, and 1.5 m.

[0867] XI.(c) Maximum overtaking speed—stopped vehicles—local

[0868] The maximum local road overtaking speed when traveling within the preferred lateral distance of the abnormally stopped vehicle boundary area may be a predetermined threshold speed below the speed limit. The preferred lateral distance may be a predetermined lateral distance.

[0869] The predetermined threshold speed may be an adjustable parameter, with a nominal value being any one of 8 mph, 10 mph, 12 mph, 15 mph, 18 mph and 20 mph.

[0870] XI.(d) Lane change reaction distance for stopped vehicles

[0871] If a lane change is required for an abnormally stopped vehicle boundary area, the automated driving vehicle may respond no later than the distance required to successfully change lanes before reaching the boundary area or a predetermined threshold minimum distance, whichever is greater.

[0872] The predetermined threshold minimum distance may be an adjustable parameter, and a nominal value thereof may be any one of 200 m, 250 m, 275 m, 300 m, 325 m or 350 m.

[0873] XI.(e) Preferred lateral distance from stopped vehicle

[0874] The automated vehicle may tend to drive at a lateral distance of at least a predetermined threshold distance measured from the widest point of the automated vehicle combination to the widest point of the abnormally stopped vehicle boundary area. The predetermined minimum lateral distance may be an adjustable parameter with a nominal value of any one of 8 feet, 10 feet, 12 feet, 14 feet, or 15 feet.

[0875] XI.(f) Slowing down and passing strategies for stopped vehicles

[0876] When the preferred lateral distance to the boundary area of ​​the abnormally stopped vehicle cannot be met, the autonomous driving vehicle may decelerate and pass through the boundary area within the preferred lateral distance only when a collision can be avoided.

[0877] XI.(g) Lane change priority - lane does not intersect with STV

[0878] The lane change priority when within a predetermined threshold distance but not in a lane crossed by an abnormally stopped vehicle boundary area may be non-critically safe. The predetermined minimum lateral distance may be an adjustable parameter, with a nominal value of any one of 8 feet, 10 feet, 12 feet, 14 feet, or 15 feet.

[0879] XI.(h) Lane change priority - lane intersecting with STV

[0880] Lane change priority when the lane is crossed by an abnormally stopped vehicle boundary area can be critical for safety.

[0881] XI.(i) Stopping emergency vehicles

[0882] If the abnormally stopped vehicle boundary area includes an emergency vehicle, the autonomous vehicle may follow the requirements in the “Emergency Vehicle Decision” subsection of this patent document.

[0883] XI.(j) Do not overtake stopped vehicles

[0884] When it is impossible to drive in a lane that is not crossed by an abnormally stopped vehicle boundary area, the autonomous driving vehicle can stop completely before reaching the boundary area and perform an MRC maneuver after a predetermined number of seconds have passed. This type of behavior can be an option to change lanes or overtake a stopped vehicle.

[0885] If the autonomous vehicle is operating in a lane that is crossed by an abnormally stopped vehicle and the autonomous vehicle cannot change lanes, the autonomous vehicle may come to a complete stop before reaching the stopped vehicle and the operations team (e.g., a chase vehicle operator), the supervisory system, or a remote control operator may issue a command to cause the autonomous vehicle to perform a minimum risk maneuver (e.g., an MRC instruction).

[0886] XI.(k) Maximum overtaking speed—stopped vehicle—highway

[0887] The maximum highway overtaking speed when driving within a preferred lateral distance of the abnormally stopped vehicle boundary zone may be an adjustable parameter having a predetermined nominal value below the speed limit.

[0888] The maximum highway overtaking speed when driving within the preferred lateral distance of the abnormally stopped vehicle boundary area can be a predetermined speed lower than the speed limit. The predetermined speed can be an adjustable parameter, and its nominal value is any one of 15mph, 18mph, 20mph, 22mph or 25mph.

[0889] Fig. 20 An example flow chart of an autonomous driving operation performed by a vehicle to operate on a road with stopped vehicles is shown. Operation 2002 includes: obtaining an image from a camera located on the autonomous vehicle by a computer located in the autonomous vehicle, wherein the image represents an area toward which the autonomous vehicle is driving on the road. Operation 2004 includes: performing a first determination based on the image to determine that the vehicle is stopped in the area for reasons unrelated to traffic congestion, traffic signals, or traffic signs. Operation 2006 includes: performing a second determination to determine that the autonomous vehicle is expected to be driven within a predetermined lateral distance from the vehicle. Operation 2008 includes: in response to the first determination and the second determination, operating the autonomous vehicle to automatically operate the vehicle at a speed less than the maximum speed allowed for the autonomous vehicle to exceed or overtake the vehicle. In some embodiments, operating the autonomous vehicle to operate at a speed less than the maximum speed allowed includes sending an instruction to an actuator in a brake unit to apply a brake or sending an instruction to an engine to reduce the speed.

[0890] In some embodiments, the method further includes: operating the autonomous vehicle to turn from the first lane to a second lane adjacent to the first lane, the distance from the vehicle being greater than or equal to the predetermined lateral distance from the vehicle. In some embodiments, the method further includes: performing a third determination to determine that the vehicle is stopped in the same lane as the autonomous vehicle; performing a fourth determination to determine that the autonomous vehicle cannot change lanes; and in response to the third determination and the fourth determination, operating the autonomous vehicle to apply braking to stop the autonomous vehicle. In some embodiments, the maximum speed is based on whether the autonomous vehicle is operating on a local road or highway.

[0891] XII Speed ​​Control

[0892] XII.(a) Speed ​​Control – Obey Speed ​​Limits

[0893] Automated vehicles can always drive at or below the posted speed limit.

[0894] XII.(b) Speed ​​Control - Observance of Speed ​​Limits - Night Speed

[0895] When driving after the sun sets, autonomous vehicles can adhere to any night-time specific speed limits.

[0896] XII.(c) Speed ​​Control – Complying with Contractual Speed ​​Limits

[0897] An autonomous vehicle may always drive at or below any existing contractual speed limit.

[0898] A contractual speed limit is a limit set on the maximum speed of an autonomous vehicle system, which is typically specified by the terms of a contract with a partner or agreed upon by a group of stakeholders.

[0899] XII.(d) Speed ​​Regulation of Autonomous Vehicle Control

[0900] The autonomous vehicle may maintain the posted speed limit (or lower) if control requires a reduction from the current speed. For example, when the autonomous driving system determines that weather conditions or road conditions do not allow the autonomous vehicle to operate at the posted speed limit because the autonomous vehicle would be in danger of losing control or there is not enough distance between it and the NPC vehicle ahead, then the current speed of the autonomous vehicle may be reduced from the posted speed limit.

[0901] XII.(e) Detect all speed limit signs

[0902] The autonomous vehicle is capable of detecting and classifying all speed limit signs, including those on local roads, highways, construction zones, and on- and off-ramps. Such detection and classification may be performed by the autonomous vehicle using data acquired by a sensor suite on the autonomous vehicle and a computing module on the autonomous vehicle configured to identify speed limit signs based on any of: sign color, overall sign shape, and a reading of an icon or text on the sign. Alternatively or additionally, a map or map database may have identified areas of changed speed limits, or identified construction zones or other types of temporary speed limit changes, and the autonomous driving system may be more vigilant in these areas to evaluate signs of speed limit postings.

[0903] XII.(f) Speed ​​Limit Increase – Maximum Acceleration

[0904] When approaching an increase in the speed limit, the autonomous vehicle may proactively accelerate to the target speed using an acceleration rate below a predetermined threshold. The predetermined threshold for the acceleration rate may be optimized for best fuel efficiency unless the autonomous vehicle is behind schedule and route arrival performance needs to be prioritized.

[0905] XII.(g) Limit acceleration and deceleration according to control needs

[0906] At all speed adjustments, the autonomous vehicle may set limits on acceleration and deceleration to ensure that the tractor and trailer do not shake and tip, sway, or slide. The autonomous vehicle may determine its own orientation using sensors including one or more inertial measurement units (IMUs), data acquired by cameras and other sensors, etc., to determine not only the current orientation of the autonomous vehicle, but also to predict possible changes in the orientation of the autonomous vehicle based on possible loss of control due to changes in the speed of the autonomous vehicle.

[0907] XII.(h) Speed ​​Limit Reduction - Engine Braking Preferred

[0908] When approaching a reduction in speed limit, the autonomous vehicle may actively decelerate to the target speed using only engine braking, unless additional deceleration is required for an evasive maneuver. Engine braking may be implemented in an autonomous vehicle with an internal combustion engine by employing any of J-brakes, cylinder deactivation, or a downshift in the transmission.

[0909] XII.(i) Uphill – Power regulation under load

[0910] When driving uphill, the autonomous vehicle can provide additional power as needed to maintain the target speed with varying trailer loads.

[0911] XII.(j) Preventive Speed ​​Slowdown – T-Intersection

[0912] When traveling in a through lane that is directly perpendicular to a non-through lane at a T-intersection, if there is a vehicle stopped or approaching in the non-through lane, the automated driving vehicle may perform a preventative deceleration (engine braking only) of no more than a predetermined amount of mph below the speed limit (such as 5 mph below the speed limit, 10 mph below the speed limit, 15 mph below the speed limit, including 20 mph below the speed limit).

[0913] XII.(k) Corners and Turns – Post-apex Behavior

[0914] When turning on a curve or intersection, the autonomous vehicle can accelerate at a ramp value after passing the apex of the curve / turn to ensure smooth acceleration and deceleration.

[0915] XII.(l) Preventive Speed ​​Slowdown—Signalized Intersections

[0916] When approaching a signalized intersection, the autonomous driving vehicle can begin preventive deceleration at a predetermined distance before the intersection (such as 90m from the intersection, 100m from the intersection, 110m from the intersection, including 120m from the intersection).

[0917] The maximum passing speed of an automated vehicle at a predetermined distance from an intersection may be equal to the posted speed limit or up to 50 mph.

[0918] XII.(m) Maps - Update speed limits

[0919] When an autonomous vehicle encounters a new speed limit sign or a speed limit sign with an updated limit, the map used by the autonomous vehicle can update the speed limit information.

[0920] XII.(n) Maps - Speed ​​Limit Information

[0921] Maps used by self-driving vehicles could include speed limit information for all mapped routes.

[0922] XII.(o) Engine Braking Efficiency

[0923] When seeking a gap to change lanes, the autonomous vehicle may be inclined to use engine braking for efficiency lane change intentions or lower priority intentions.

[0924] XII.(p) Downhill - Engine braking preferred

[0925] When going downhill and needing to slow down, an autonomous vehicle may be inclined to use engine braking.

[0926] XII.(q) Corners and Turns – Pre-apex Behavior

[0927] When approaching or turning at a curve or intersection, the automated vehicle may decelerate at a ramp value to ensure smooth acceleration and deceleration before reaching the apex of the curve, preferably using engine braking to decelerate.

[0928] XII.(r) Speed ​​Control—Observe Speed ​​Limits—Ramps

[0929] Automated vehicles may comply with any speed limits posted on entrance or exit ramps when merging onto or off a highway.

[0930] XII.(s) Speed ​​Limit Signs Road Association

[0931] Automated vehicles can associate speed limit signs with the correct road structure (e.g., ramp speed limit vs. highway speed limit). For example, a speed limit sign on an exit ramp but still visible from the highway can be associated with the ramp instead of the highway.

[0932] XII.(t) Supervision – Update rate limits

[0933] When encountering a new speed limit sign or a speed limit sign with an updated limit, the autonomous vehicle may transmit the information to the supervisory system, including a remote control operator associated with the supervisory system, who may be responsible for communicating the updated speed limit information to the rest of the fleet.

[0934] XII.(u) Speed ​​Control – Speed ​​Limit Timing

[0935] The automated vehicles may be at or below the speed limit until the frontmost point of the automated vehicle combination reaches the speed limit sign.

[0936] XII.(v) Engine braking for following distance

[0937] An autonomous vehicle may be inclined to use engine braking when increasing or maintaining a following distance gap to another vehicle.

[0938] XII.(w) Speed ​​control - obeying speed limits - night speed

[0939] When driving after the sun sets, autonomous vehicles can adhere to any night-time specific speed limits.

[0940] XII.(x) Maps - Speed ​​Limit Information

[0941] Maps used by self-driving vehicles could include speed limit information for all mapped routes.

[0942] XII.(y) Engine braking for efficient lane change

[0943] When seeking a gap to change lanes, an automated vehicle may be inclined to use engine braking for efficiency lane change intent or lower priority intent.

[0944] XII.(z) Downhill - Engine braking preferred

[0945] When going downhill and needing to slow down, an autonomous vehicle may be inclined to use engine braking.

[0946] XII.(aa) Speed ​​Limit Increase – Optimum Acceleration

[0947] After passing a speed limit sign with an increased limit, the autonomous vehicle may accelerate to the target speed using an acceleration rate optimized for best fuel efficiency, unless the autonomous vehicle is behind schedule and route arrival performance needs to be prioritized.

[0948] In some embodiments, a supervisory system including a remote control operator associated with the supervisory system can provide guidance when greater acceleration is required.

[0949] XII.(aa) Speed ​​Control When Approaching a Signalized Intersection

[0950] When approaching a local signalized intersection, the autonomous vehicle may reduce its speed based on a maximum speed equal to the posted speed limit or up to 50 mph. The autonomous vehicle may reach a target speed at least a predetermined distance before the intersection (such as 90 m from the intersection, 100 m from the intersection, 110 m from the intersection, including 120 m before the stop line of the intersection, as measured from the front bumper of the autonomous vehicle to the stop line).

[0951] An autonomous vehicle may prefer to use engine braking or coasting to achieve the desired deceleration.

[0952] XIII Slope

[0953] XIII.(a) Braking control

[0954] An autonomous vehicle may prioritize the use of engine brakes over foundation brakes (e.g., disc brakes, drum brakes for each axle or each wheel) to maintain the effectiveness of the foundation brakes and prevent the foundation brakes from overheating.

[0955] XIII.(b) Slope road detection

[0956] The autonomous vehicle can use onboard sensors to detect when the autonomous vehicle is driving on a sloping road based on the slope. Alternatively or additionally, information about changes in road slope can be marked on a map utilized by the autonomous vehicle, and the position of the autonomous vehicle in conjunction with the mapped data can confirm detection of changes in vehicle orientation corresponding to the road slope.

[0957] XIII.(c) Out-of-control ramp

[0958] In the event of brake weakening or failure, the autonomous vehicle can slow down and stop using the runaway ramp.

[0959] XIII.(d) Slope road sign recognition

[0960] Autonomous vehicles are able to recognize signs indicating a sloped road.

[0961] XIII.(e) Engine Braking Inhibit

[0962] If a “No Engine Braking” sign is encountered, the automated vehicle must not engage the engine brake within a minimum predetermined threshold distance.

[0963] XIII.(f) Lane Change Avoidance

[0964] When driving on a sloped road, the autonomous vehicle can avoid all types of efficiency lane changes and lower priority lane changes.

[0965] XIII.(g) Speed ​​Control

[0966] When driving on a sloped road, the autonomous vehicle can select an appropriate speed to prevent the trailer from tipping, swaying or sliding.

[0967] The autonomous vehicle may take into account the steepness of the grade, the curvature of the road, road traction conditions, prevailing weather conditions, visibility conditions, and the weight and center of gravity of the autonomous vehicle and trailer.

[0968] XIII.(h) Contacting the Operator

[0969] If the autonomous vehicle is on an off-ramp and has come to a stop, the autonomous vehicle can remain stationary and contact the operator.

[0970] XIII.(i) Occupied uncontrolled ramp

[0971] When the runaway ramp is occupied by another vehicle, the autonomous vehicle may still use the runaway ramp, but will veer away to avoid the vehicle already on the runaway ramp.

[0972] XIII.(j) Drawing

[0973] Autonomous vehicles can map sloped roads and known runaway ramps for navigation purposes.

[0974] XIII.(k) Scrolling Backwards

[0975] Automated driving can prevent a vehicle from rolling backwards when stopping or starting from a stop on a sloped road.

[0976] XIII.(l) Out-of-control ramp sign

[0977] Autonomous vehicles are able to identify runaway ramps based on pre-mapped locations and road signs.

[0978] XIII.(m) Right Lane Preference

[0979] When on a sloped road, the automated vehicle may drive in the right lane unless it is to make an evasive maneuver or avoid an ELV (e.g., a disabled or incapacitated vehicle).

[0980] XIII.(n) Description of Slope Road

[0981] A sloped road may be defined as a road having a slope exceeding a predetermined amount (eg, 2%, 3%, 5%, 6%, etc.).

[0982] XIII.(o) Out-of-control procedures

[0983] When the automated vehicle is unable to stop due to brake reduction or failure, the automated vehicle may engage maximum engine braking, turn on hazard lights, and use the horn to warn other road users. The automated vehicle may change lanes to the rightmost lane at the first possible opportunity to enable use of the runaway ramp when available.

[0984] XIII.(p) Hazard lights

[0985] Unless in a traffic jam, an autonomous vehicle may engage its hazard lights on a sloped road if the autonomous vehicle is driving below the speed limit by more than a predetermined threshold amount (e.g., 10mph, 15mph, 20mph, 25mph, 30mph, etc.) or if the autonomous vehicle is driving at a speed less than a predetermined threshold level (e.g., 35mph, 40mph, 45mph, 50mph, etc.).

[0986] XIV. Highway Exits

[0987] The compliance module of the onboard control computer of the autonomous vehicle can perform image processing on the traffic sign to identify the information indicated by the traffic sign, as further explained in this section.

[0988] XIV. (a) Merging at a blocked highway exit

[0989] When a highway exit is blocked and the autonomous vehicle cannot queue at the end, the autonomous vehicle may slow down to no slower than a predetermined threshold amount below the average highway traffic speed to seek a gap to merge in. If the autonomous vehicle cannot merge, the autonomous vehicle may use an alternate route.

[0990] XIV.(b) Blocked Highway Exits

[0991] When a highway exit is blocked, the autonomous vehicle is able to detect the NPC queue on the exit lane and slow down to join at the end of the queue.

[0992] XIV.(c) Stay in the exit lane

[0993] The automated driving vehicle may remain in the exit lane within a minimum predetermined distance before the exit point. The minimum predetermined distance before the exit point may be approximately 800 m (0.5 mile), 1200 m (0.75 mile), 1600 m (1 mile), or 2000 m (1.25 miles).

[0994] XIV.(d) Multi-lane exits

[0995] An autonomous vehicle at a multi-lane exit can choose to drive in the lane that is best for the next leg of its journey after leaving the exit ramp.

[0996] XIV.(e) Closed Exits

[0997] When an exit that the autonomous vehicle intends to use is closed, the autonomous vehicle can rejoin the highway and use an alternate route.

[0998] XIV.(f) Export Zone—Description

[0999] The exit area may be defined as the area beginning at the location where the exit ramp lane begins to separate from the highway and ending at the point where the exit ramp is separated from the highway by a solid line, a lane divider, a hard shoulder, or a soft shoulder.

[1000] XIV.(g) Alternative routes

[1001] The autonomous vehicle can plot alternative routes as backups to all highway exits to ensure that the autonomous vehicle will eventually reach its destination.

[1002] XIV.(h) Dedicated exit lane

[1003] When a dedicated exit lane is available for an exit, the automated vehicle may change lanes to the exit lane at the first possible opportunity.

[1004] XIV.(i) Highway exits during traffic jams

[1005] In traffic jams, autonomous vehicles can identify and enter exit lanes by creating gaps in order to merge into the exit lane.

[1006] XIV.(j) Drawing

[1007] Autonomous vehicles can identify and plot highway exits and alternative routes in navigation maps.

[1008] XIV.(k) Export Area

[1009] Autonomous vehicles are able to identify exit areas on highways to determine where to exit.

[1010] XIV.(l) Exit sign recognition

[1011] Self-driving vehicles are able to identify highway exits based on signs.

[1012] XIV.(m)k-Ramp Exit

[1013] The autonomous vehicle may seek a gap when exiting the k-ramp highway exit. If the autonomous vehicle cannot exit due to insufficient gap, the autonomous vehicle may use an alternate route.

[1014] XIV.(n) Exit Ramp Speed ​​Limit

[1015] The autonomous vehicle may drive at a speed lower than the exit ramp speed limit throughout the exit ramp.

[1016] XIV.(o) Speed ​​reduction

[1017] The autonomous vehicle can decelerate to a safe speed for the exit ramp before exiting the highway exit and avoid significant deceleration beyond a predetermined threshold to prevent tipping and swaying of the trailer.

[1018] XV. Go straight

[1019] XV.(a) Lane selection

[1020] The autonomous vehicle may identify and select a lane to proceed straight through the intersection based on the lane from which the autonomous vehicle initiated the crossing.

[1021] XV.(b) Traffic light occlusion

[1022] If the traffic light is obscured, the autonomous vehicle can stop before the intersection and then slow forward to get a better view of the traffic light. If the traffic light is still obscured, the autonomous vehicle can treat the intersection as an unprotected TTC stop.

[1023] XV.(c) Cross-traffic conditions

[1024] Automated driving vehicles can take into account cross traffic conditions and give way to non-compliant cross traffic vehicles that do not stop.

[1025] XV.(d) Faulty traffic lights – stop sign

[1026] If the traffic light is off or flashing red, the autonomous vehicle can treat the intersection as a stop sign intersection.

[1027] XV.(e) Avoidance of hard deceleration

[1028] When braking for a yellow light, the automated vehicle may avoid hard deceleration exceeding a predetermined threshold. The predetermined threshold deceleration value may not exceed 3.8 m / s 2 , not more than 3.4m / s 2 , such as not more than 3.0m / s 2 , including no more than 2.5m / s 2 .

[1029] If, when the traffic light turns yellow, the autonomous vehicle determines that the autonomous vehicle will need to decelerate beyond a predetermined threshold in order to stop prior to the intersection, the autonomous vehicle may accelerate just enough to cross the intersection and ensure that a portion of the autonomous vehicle is in the intersection before the yellow light interval ends.

[1030] XV.(f) Lane change avoidance

[1031] When an autonomous vehicle is within an intersection, the autonomous vehicle may avoid lane changes unless it is for an evasive maneuver.

[1032] XV.(g) Traffic light detection

[1033] The autonomous vehicle is able to identify traffic lights and determine when the traffic light signal in the autonomous vehicle's travel path indicates that the autonomous vehicle is allowed to cross the intersection.

[1034] XV.(h) Faulty traffic lights - give way

[1035] When an autonomous vehicle encounters a faulty traffic light that is flashing yellow, the autonomous vehicle can avoid stopping and instead appropriately slow down according to local regulations and give way to an NPC already in the intersection.

[1036] XV.(i) Intersection Blockage

[1037] The autonomous vehicle may avoid entering the intersection if the portion of the intersection that the autonomous vehicle is about to traverse is still occupied by traffic.

[1038] XV.(j) Yellow Light Behavior

[1039] If the traffic light in the intersection turns yellow before the autonomous vehicle reaches the intersection, the autonomous vehicle can determine whether the autonomous vehicle can cross the intersection based on the time required to reach and cross the intersection.

[1040] XV.(k) Intersection spacing

[1041] If there is not enough room outside the intersection for the autonomous vehicle to fully pass through the intersection, the autonomous vehicle may avoid entering the intersection.

[1042] XV.(l) Drawing

[1043] Autonomous vehicles can identify and map intersections for navigation purposes.

[1044] XV.(m) Duration of yellow light

[1045] If the duration of the yellow light is unknown, the autonomous vehicle can assume that the duration of the yellow light is a predetermined number of seconds.

[1046] XV.(n) Faulty traffic lights

[1047] If the autonomous vehicle detects that a light at an intersection is not operating in a normal manner, the autonomous vehicle may classify the traffic light as faulty.

[1048] Types of faulty traffic lights can include:

[1049] 1. Flashing red light

[1050] 2. Flashing yellow light

[1051] 3. Traffic lights outage

[1052] XV.(o) Blocking the intersection

[1053] The automated vehicle must not enter the intersection when the automated vehicle predicts that it will not completely exit the intersection before the end of the red light passing interval. The red light passing interval may be defined as the amount of time between when a signal light at the intersection changes from green to red in a first direction and includes the time before it changes from red to green in a second direction that intersects the first direction.

[1054] XVI. Object Perception and Attribute Analysis

[1055] XVI. (a) Object Perception

[1056] An autonomous vehicle can detect and analyze properties of objects within its field of view (FOV). The field of view can be defined for each camera or sensor on the autonomous vehicle and for the autonomous vehicle as a whole.

[1057] The autonomous vehicle can associate objects with the lane or shoulder it is in (OTLA: Object to Lane Association). For example, a vehicle parked in the emergency lane or shoulder can be classified as an Emergency Lane Vehicle (ELV), while a vehicle in an entrance ramp or merging lane can be classified as a merging vehicle.

[1058] The autonomous vehicle may classify objects as static or moving. The classification may be performed by a computing module of the autonomous driving system of the autonomous vehicle, which analyzes sensor data from different time periods to compare the data and the relative positions of any objects in the data, thereby explaining changes in the position of the autonomous vehicle, etc.

[1059] The autonomous vehicle may have the ability to reclassify an object as its behavior changes. For example, an NPC vehicle may change from moving to static (i.e., a stopped vehicle) from one moment to the next, while debris in a lane may change from static to moving once struck by a vehicle traveling in the lane (e.g., an NPC vehicle).

[1060] XVI.(b) Static Objects

[1061] An autonomous vehicle can detect static objects within its FOV and road boundaries and analyze their properties.

[1062] The autonomous vehicle can detect static objects with a predetermined minimum height (e.g., 1 inch, 2 inches, 3 inches, 4 inches, 5 inches).

[1063] An autonomous vehicle can detect all static objects that affect its capabilities (e.g., perception, localization, behavior, and driving capabilities).

[1064] An autonomous vehicle may have the ability to detect and analyze attributes of static objects, including any one or more of: size (dimensions); state as immovable or movable (expected to move); and classification (stopped vehicle, traffic sign, traffic light, guardrail, bridge, etc.).

[1065] An autonomous vehicle may detect static objects within a predetermined minimum distance depending on and relative to the size of the object.

[1066] XVI.(c) Moving / Movable Objects

[1067] An autonomous vehicle can detect moving objects within its FOV and road boundaries and analyze their properties.

[1068] The autonomous vehicle may have the ability to detect and analyze properties of static objects, including any one or more of the following: size (dimensions); the autonomous vehicle may detect the width and height of a moving object from at least a first predetermined number of meters away, and its length from at least a second predetermined number of meters away; speed; direction; acceleration / deceleration; position (relative to the position of the autonomous vehicle); and associated lane (or shoulder).

[1069] XVII. Emergency Vehicles

[1070] XVII.(a) Description of flash emergency vehicles

[1071] With respect to the detection of autonomous vehicles, a flashing emergency vehicle can be characterized as a vehicle having at least one headlight that exhibits red light or red-blue light. Vehicles with flashing lights, including tow trucks, construction vehicles, fire trucks, ambulances, and police cars, can all be considered flashing emergency vehicles.

[1072] XVII.(b) Flashing emergency vehicle (EV) detection

[1073] The autonomous vehicle may react to the following flashing emergency vehicles (EVs) no later than a predetermined number of meters (e.g., 125m, 150m, 152.4m, 175m, 200m) before the arrival of an emergency vehicle (EV), unless the flashing EV is initially obscured and cannot be detected. The types of flashing EVs that require a reaction from the autonomous vehicle include: a flashing EV that is in front of the autonomous vehicle and has come to a complete stop; a flashing EV that is in front of the autonomous vehicle and is moving in the same direction of travel; a flashing EV that is behind the autonomous vehicle and is moving in the same direction of travel; and a flashing EV that is behind the autonomous vehicle and is moving in the opposite direction of travel on the same roadway.

[1074] A carriageway is defined as the width of roadway over which vehicles can move laterally without being restrained by any physical barriers or divisions.

[1075] XVII.(c) Flash EV - Memory

[1076] The autonomous vehicle may save in memory the presence of a flashing emergency vehicle for a minimum predetermined number of seconds (e.g., 10s, 12s, 15s, 18s) that is later fully or partially obscured from view.

[1077] XVII.(d) Flash EV – Lane Change Priority

[1078] Any lane change attempt or refusal caused by a flashing emergency vehicle may be prioritized as critical safety, ie, the highest priority.

[1079] XVII.(e) Flashing EV - Stopping on the Highway - Lane Changing Strategy

[1080] For a flashing emergency vehicle that is stopped on the highway (on the shoulder or off-shoulder area), the automated vehicle may leave at least one full lane between itself and the emergency vehicle unless a lane change is not possible or denied.

[1081] Stopping on the highway - Lane change strategy - Maximum overtaking speed:

[1082] When the autonomous vehicle has at least one full lane between itself and a flashing emergency vehicle stopped on the highway, the autonomous vehicle may overtake the EV at a maximum speed that is a predetermined amount of mph below the speed limit (e.g., 8mph, 10mph, 12mph).

[1083] Stopping on the highway - Lane change strategy - Maximum overtaking speed - Nearby traffic:

[1084] When the autonomous vehicle has at least one full lane between itself and a flashing emergency vehicle stopped on the highway, the autonomous vehicle may pass the EV at a speed that is less than a predetermined mph amount (e.g., 5mph, 8mph, 10mph) above the average speed of non-emergency, non-stationary vehicles within a predetermined distance (e.g., 125m, 150m, 175m, 200m) of autonomous vehicles traveling in the same direction on the road.

[1085] Stopped on the Highway – Lane Change Intent Timing:

[1086] The lane change intention of the autonomous vehicle, where the lane change intention is associated with a stopped flashing emergency vehicle (EV), can preferably remain effective from the time the flashing EV is detected until the time the front bumper of the autonomous vehicle passes the rear bumper of the EV.

[1087] Stopped on the highway - Lane change rejection timer:

[1088] A lane change rejection associated with a stopped flashing emergency vehicle (EV) may preferably remain in effect from the time the flashing EV is detected until the time the rearmost point of the autonomous vehicle combination (including the trailer) passes the front bumper of the EV.

[1089] Stopping on the highway - Lane changing strategy - Multiple EVs:

[1090] When there are multiple flashing emergency vehicles parked on a highway, the automated vehicle may leave at least one full lane between itself and the emergency vehicle that protrudes farthest into the road, unless a lane change is impossible or denied.

[1091] Stopping on the highway - multiple EVs - lane change rejection timing:

[1092] A lane change rejection associated with a stopped multiple flashing emergency vehicles (EVs) may preferably remain in effect from the time the flashing EV is detected until the time the rearmost point of the autonomous vehicle combination (including the trailer) passes the front bumper of the last EV.

[1093] XVII.(f) Flash EV - Stopped on the Highway - Deceleration Strategy

[1094] When it is not possible to pass a complete lane between itself and a flashing emergency vehicle parked on the highway (on the shoulder or off-shoulder area), the autonomous vehicle may preferably slow down according to the "Pedestrian Slowdown Strategy" described above, unless local regulations dictate otherwise.

[1095] The lateral distance in the “Pedestrian Deceleration Strategy” can be measured from the widest point of the autonomous vehicle to the widest point of the EV.

[1096] XVII.(g) Flashing EV - Moving on the shoulder - Lane change strategy

[1097] For a flashing emergency vehicle moving on the shoulder of a highway, the autonomous vehicle may preferably follow the above-mentioned behavioral strategy, as described in the section entitled "Flashing EV - Stopped on Highway - Lane Changing Strategy".

[1098] XVII.(h) Flashing EV - Moving on the shoulder - Alternative strategy

[1099] When it is unable to pass a full lane between itself and a flashing emergency vehicle moving on the shoulder of a highway, the autonomous vehicle may slow down in accordance with the behaviors described above in the section entitled “Flashing EV—Stopped on Highway—Slowdown Strategy.”

[1100] Flash EV - Moving on the shoulder - Anticipating the cut:

[1101] When a full lane cannot be passed between itself and a flashing emergency vehicle moving on the shoulder of a highway, the autonomous vehicle may predict whether a flashing EV will cut in front of the autonomous vehicle as outlined below. When a full lane cannot be passed between itself and a flashing emergency vehicle stopped on a highway (shoulder or off-shoulder area), the autonomous vehicle may slow down and maneuver so that there may be at least one lane between the autonomous vehicle and vulnerable pedestrians around the flashing EV, and when such vulnerable pedestrians are detected, and if achieving 1 lane is not possible, the autonomous vehicle may slow down and offset away from the vulnerable pedestrian while maintaining the widened lane, unless local regulations dictate otherwise.

[1102] Flash EV - Moving on the shoulder - Accepting the cut:

[1103] When determining whether to pass a flashing emergency vehicle moving on the shoulder of the road or allow it to cut in, the autonomous vehicle may behave as follows. When the autonomous vehicle detects that the cutting-in vehicle is traveling faster than the autonomous vehicle, the autonomous vehicle may reduce the throttle to achieve deceleration. When the autonomous vehicle detects that the cutting-in or merging vehicle is traveling slower than the autonomous vehicle, the autonomous vehicle may choose to change lanes to the left to accommodate the cutting-in or merging vehicle.

[1104] XVII.(i) Flash EV – EV Approaching Autonomous Driving Vehicles

[1105] When a flashing emergency vehicle is approaching the autonomous vehicle from behind at an equal or greater speed and in the same direction of travel (highway or local), the autonomous vehicle may yield the right of way to the emergency vehicle.

[1106] Yield to EVs - Intersection constraints:

[1107] When slowing down and yielding to flashing emergency vehicles, automated vehicles cannot stop in the middle of an intersection.

[1108] EV approaching autonomous vehicle - both in the rightmost lane:

[1109] Figure 4A An example scenario is shown of an EV 402 approaching an autonomous vehicle 404 on a road. If the autonomous vehicle is in the rightmost lane, and a flashing emergency vehicle is approaching the autonomous vehicle and is also in the rightmost lane, the autonomous vehicle can use any available paved section of the shoulder to position itself to the right (outside the lane, if possible), and can gradually slow down until the EV passes.

[1110] Figure 4B and Figure 4C An example scenario is shown of an autonomous vehicle with EV 402 approaching the right side of autonomous vehicle 404. If a flashing emergency vehicle is approaching the autonomous vehicle, and the EV is in one or more lanes to the right of the autonomous vehicle, the autonomous vehicle may remain in its current lane and gradually decelerate until the EV passes.

[1111] EV approaches autonomous vehicle - moves to rightmost lane:

[1112] If a flashing emergency vehicle is approaching an autonomous vehicle and is not FIG. 4A to FIG. 4C In the above situation (e.g., a situation where both the EV and the autonomous vehicle are in the rightmost lane, or a situation where the EV approaches the autonomous vehicle to the right of the autonomous vehicle), the autonomous vehicle may position itself in the rightmost lane and may gradually decelerate until the emergency vehicle passes. For example, FIG. 5A to FIG. 5CAn example scenario is shown in which an emergency vehicle 402 approaches from the left side of the autonomous vehicle 404 on the road, or an example scenario in which neither the emergency vehicle nor the autonomous vehicle is in the rightmost lane of the road. FIG. 5D to FIG. 5E An example scenario is shown where emergency vehicle 402 approaches autonomous vehicle 404 , where neither emergency vehicle 402 nor autonomous vehicle 404 is in the rightmost lane on the road.

[1113] EV approaching autonomous vehicle – lane change intention and rejection timing:

[1114] The critical safety lane change intention(s) or rejections associated with a flashing EV approaching the autonomous vehicle may preferably remain in effect from the point in time when the flashing EV is detected until the point in time when the rear bumper of the EV passes the front bumper of the autonomous vehicle.

[1115] XVII.(j) Flashing EV – The autonomous vehicle follows the EV in the same direction of travel

[1116] When the autonomous vehicle is behind a moving, flashing emergency vehicle traveling on the same roadway in the same direction of travel, the autonomous vehicle cannot pass the emergency vehicle unless the emergency vehicle is transitioning to a stopping EV (e.g., a stationary vehicle).

[1117] Autonomous vehicle following EV - maintaining longitudinal distance:

[1118] When the autonomous vehicle is behind a moving flashing emergency vehicle traveling on the same roadway in the same direction of travel, the autonomous vehicle may maintain a predetermined longitudinal distance from the EV unless it is transitioning to a stopping EV (e.g., a stationary vehicle).

[1119] Autonomous vehicle following EV – transition to stop EV:

[1120] If a moving flashing emergency vehicle meets any of the following criteria, the automated vehicle may assume that the vehicle is transitioning to a stop (e.g., a static object) and may follow an "EV stopped on highway" strategy to pass the vehicle. The criteria may include any of the following: the flashing EV is in the leftmost lane, rightmost lane, shoulder, or lane divider of a multi-lane roadway, is traveling below the speed limit by more than a predetermined amount of mph (e.g., 25mph,...

Claims

1. A method of operating an autonomous vehicle, comprising: acquiring, by a computer located in the autonomous vehicle, an image from a camera located on the autonomous vehicle, wherein the image represents an area toward which the autonomous vehicle is driving on a lane on a road or highway; determining, based on the image, that a pedestrian or a cyclist is located beside the lane on the road or highway; In response to the determination, operating the automated driving vehicle to turn from the center of the lane to a first side of the lane, the first side of the lane being away from the center of the lane and away from the location of the pedestrian or the cyclist; as well as In response to the determination, the autonomous vehicle is operated to reduce the speed of the autonomous vehicle to below a first threshold speed value in response to determining that the lateral distance from the autonomous vehicle to the pedestrian or the cyclist is within a first distance set and that the current speed of the autonomous vehicle is greater than a first threshold speed value.

2. The method according to claim 1, wherein the autonomous vehicle is caused to reduce the speed of the autonomous vehicle by comparing the lateral distance from the autonomous vehicle to the pedestrian or the cyclist and the current speed of the autonomous vehicle to a table comprising a plurality of distance sets and a plurality of threshold speed values, wherein the plurality of distance sets include the first distance set and a second distance set that is greater than or equal to the first distance set, wherein the plurality of threshold speed values ​​include the first threshold speed value and a second threshold speed value greater than the first threshold, and The first distance set and the second distance set correspond to the first threshold speed value and the second threshold speed value respectively.

3. The method according to claim 2, The first threshold speed value is the minimum value between the first predetermined speed value and the first speed value, wherein the first speed value is obtained by subtracting a specific speed from a speed limit, and The second threshold speed value is a minimum value between a second predetermined speed value and the first speed value.

4. The method according to claim 2, further comprising: In response to determining that the lateral distance from the autonomous vehicle to the pedestrian or the cyclist is greater than a third distance set, operating the autonomous vehicle to maintain the speed of the autonomous vehicle, wherein the third distance set is greater than or equal to the second distance set.

5. The method according to claim 2, further comprising: In response to determining the presence of an emergency vehicle on the road or the highway based on the image: operating the autonomous vehicle to reduce the speed of the autonomous vehicle below a third threshold speed value in response to determining that the lateral distance from the autonomous vehicle to the pedestrian or the cyclist is within the first set of distances and the current speed of the autonomous vehicle is greater than a third threshold speed value, The third threshold speed value is a minimum value between the first threshold speed value and a maximum overtaking speed value.

6. The method of claim 4, wherein the maximum overtaking speed value is a specific speed that is less than a speed value, and wherein the speed value is based at least on a speed limit of the road or the highway and whether the autonomous driving vehicle is operating on the road or the highway.

7. The method according to claim 1, further comprising: Operate the autonomous vehicle to pass the pedestrian or the cyclist by maintaining a minimum lateral distance between the autonomous vehicle and the pedestrian or the cyclist, wherein the minimum lateral distance is a predetermined distance from a side of the autonomous vehicle farthest from the pedestrian or the cyclist to a position of the pedestrian or the cyclist.

8. The method according to claim 1, wherein the pedestrian or the cyclist is determined based on the image when a first distance from a first position of the autonomous vehicle to a second position of the pedestrian or the cyclist is greater than or equal to a stopping distance of the autonomous vehicle, and The stopping distance is the second distance required for the autonomous vehicle to come to a complete stop.

9. A method of operating an autonomous vehicle, comprising: determining, by a computer located in the autonomous vehicle, that the autonomous vehicle is decelerating when the autonomous vehicle is at a first location on a road, the first location being within a predetermined distance of a second location at which the autonomous vehicle is to perform a steering maneuver; as well as In response to the determination and in response to determining that the turn signal lamp is not turned on, operating the turn signal lamp to be turned on at a first time.

10. The method according to claim 8, further comprising: In response to determining that the turn signal is not working or operating, sending instructions that cause the autonomous vehicle to steer along a trajectory to a side of the road and apply braking.

11. The method according to claim 8, wherein causing the turn signal to be turned on at the first time for a first time length, and The method further comprises: performing a first determination to determine that the first length of time overlaps with a second length of time associated with a second steering maneuver subsequent to the steering maneuver; performing a second determination to determine that the second steering maneuver is in the same direction as the steering maneuver; as well as In response to the first determination and the second determination, the turn signal lamp is operated to remain on during the first time length and the second time length.

12. The method of claim 8, wherein the predetermined distance is based on laws or regulations of the region or state in which the autonomous vehicle is operating.

13. The method according to claim 8, further comprising: performing a determination to determine that a second location at which the automated driving vehicle is to perform the steering maneuver is adjacent to, within a certain distance of, or past an intersection; and Wherein in response to the determination, in response to determining that the turn signal is not turned on, and in response to determining that the rear of the autonomous driving vehicle passes through the middle of the intersection, the turn signal is caused to be turned on.

14. A method of operating an autonomous vehicle, comprising: determining, by a computer located in the autonomous vehicle, that a lane of a road on which the autonomous vehicle is operating includes a curved portion having a minimum radius greater than or equal to a predetermined threshold; as well as When the autonomous driving vehicle drives through the curved portion of the road, the autonomous driving vehicle operating on the curved portion moves toward one side of the lane and away from the center of the lane, wherein the side of the lane is a side that curves outwards, and The autonomous driving vehicle is caused to move toward the side of the lane until a predetermined threshold distance from the center of the lane.

15. The method according to claim 13, further comprising: In response to determining that an additional curved portion of the road has superelevation greater than a predetermined threshold amount, operating the automated driving vehicle to avoid traveling on the additional curved portion, The superelevation describes the upward angle formed by the upward inclination of the further curved portion relative to the flat surface.

16. The method of claim 14, wherein the predetermined threshold amount is a value between 5% and 15%.

17. The method according to claim 13, further comprising: The autonomous driving vehicle is operated to accelerate or decelerate to less than or equal to a predetermined rate in a curved portion of the road.

18. The method according to claim 13, further comprising: The autonomous vehicle is operated to reduce a speed of the autonomous vehicle in response to determining that a time to collision (TTC) value when the autonomous vehicle is operating in the curved portion is greater than a predetermined amount of time, wherein the TTC value indicates an amount of time of visibility provided by one or more cameras on the autonomous vehicle.

19. A method of operating an autonomous vehicle, comprising: determining, by a computer located in the autonomous vehicle, that an emergency vehicle is within a predetermined distance of a first position of the autonomous vehicle operating in a lane on a road; as well as In response to the determination, operating the automated driving vehicle to steer from a center of the lane toward a first side of the lane, the first side of the lane being away from the center of the lane and away from a second location of the emergency vehicle, The autonomous driving vehicle is caused to turn toward the first side until a lateral distance between the emergency vehicle and the autonomous driving vehicle is greater than or equal to the predetermined distance.

20. The method of claim 18, wherein the autonomous driving vehicle is caused to turn toward a first side of the lane and onto the second lane in response to determining that a line separating the lane and a second lane adjacent to the lane comprises a white dashed line, a yellow dashed line, or a white solid line.

21. The method of claim 18, further comprising: In response to determining that the emergency vehicle is within the predetermined distance of the first position of the autonomous vehicle, and in response to determining that a lane change maneuver by the autonomous vehicle is not possible: Sending instructions that cause the autonomous vehicle to apply brakes or decelerate the autonomous vehicle to a speed less than a threshold speed value.

22. The method of claim 20, wherein the threshold speed value is based on rules of a region, state, or region in which the autonomous vehicle is located.

23. The method of claim 20, wherein the threshold is based on a speed limit of the first location at which the autonomous vehicle is operating and the lateral distance between the emergency vehicle and the autonomous vehicle.

24. The method of claim 18, wherein the autonomous vehicle operates to turn from the center of the lane toward the first side of the lane and causes the autonomous vehicle to apply braking, or decelerate the autonomous vehicle to a speed less than a threshold speed value, in response to: determining that the emergency vehicle is operating in an additional lane that is immediately adjacent to the lane on which the autonomous vehicle is operating; and Determining that a lane change maneuver by the automated driving vehicle is impossible.

25. The method of claim 18, further comprising: The autonomous vehicle is operated to accelerate only for lane change or to perform an evasive maneuver in response to determining that the emergency vehicle is approaching the autonomous vehicle and that the emergency vehicle is operating in another lane that is immediately adjacent to the lane in which the autonomous vehicle is operating.

26. A method of operating an autonomous vehicle, comprising: acquiring, by a computer located in the autonomous vehicle, an image from a camera located on the autonomous vehicle, wherein the image represents an area on a road toward which the autonomous vehicle is driving; performing a first determination based on the image to determine that the vehicle is parked in the area for a reason unrelated to traffic congestion, a traffic signal, or a traffic sign; performing a second determination to determine that the automated driving vehicle is expected to be driven within a predetermined lateral distance from the vehicle; and In response to the first determination and the second determination, the autonomous vehicle is operated at a speed that is less than a maximum speed allowed for the autonomous vehicle to overtake or exceed the vehicle.

27. The method of claim 25, further comprising: Operating the automated driving vehicle to steer from a first lane to a second lane adjacent to the first lane at a distance from the vehicle greater than or equal to the predetermined lateral distance from the vehicle.

28. The method of claim 25, further comprising: performing a third determination to determine that the vehicle is parked in the same lane as the automated driving vehicle; performing a fourth determination to determine that the autonomous driving vehicle cannot change lanes; as well as In response to the third determination and the fourth determination, the autonomous driving vehicle is operated to apply brakes to stop the autonomous driving vehicle.

29. The method of claim 25, wherein the maximum speed is based on whether the autonomous vehicle is operating on a local road or highway.

30. A method of operating an autonomous vehicle, comprising: acquiring, by a computer located in the autonomous vehicle, an image from a camera located on the autonomous vehicle, wherein the image represents an area toward which the autonomous vehicle is driving on an on-ramp of a highway; determining, based on the image, that the area includes a merging section on a lane on the highway, and that the autonomous driving vehicle is expected to merge onto the highway in the merging section; In response to the determination, operating a turn signal to turn on, wherein the turn signal indicates that the autonomous driving vehicle is expected to merge from the on-ramp into the lane on the highway; and In response to the determination and in response to the turn signal being turned on, the automated driving vehicle is operated to turn from the entrance ramp of the highway to the merging section on the lane of the highway.

31. A method according to claim 29, wherein the total length of the merging section includes the length of the autonomous driving section, a first minimum distance allowed between the autonomous driving vehicle and a first vehicle expected to be located in front of the autonomous driving vehicle, and a second minimum distance allowed between the autonomous driving vehicle and a second vehicle expected to be located behind the autonomous driving vehicle.

32. The method of claim 29, further comprising: performing a first determination to determine that the length of the merging section is decreasing; as well as In response to the first determination, the autonomous vehicle is operated to apply brakes to stop the autonomous vehicle.

33. The method of claim 29, further comprising: In response to the determination, a second determination of a trajectory followed by the automated driving vehicle from the on-ramp to the merging segment is performed, wherein the trajectory avoids the automated driving vehicle from entering a lane split angle region.

34. A method according to claim 29, wherein the image is acquired by the autonomous driving vehicle when it is determined that there are no additional merging sections in the prior image of the additional area towards which the autonomous driving vehicle is driving from the camera and when the autonomous driving vehicle is operated to move forward slowly on the highway, wherein the prior image is acquired in time before the time when the image is acquired from the camera.

35. The method of claim 33, wherein the autonomous vehicle creeps forward at a speed less than a predetermined speed.

36. A method of operating an autonomous vehicle, comprising: acquiring, by a computer located in the autonomous vehicle, a set of images over time from a first camera located on the autonomous vehicle, wherein the set of images represents an area adjacent to a lane on a road on which the autonomous vehicle is driving; acquiring, by the computer, an image from a second camera located on the autonomous vehicle, wherein the image represents an additional area of ​​the lane on which the autonomous vehicle is driving; performing a first determination based on the set of images to determine that a vehicle is driving proximate to the autonomous vehicle for a length of time; performing a second determination based on the image or the set of images to determine a level of risk associated with the autonomous vehicle driving parallel to the vehicle; performing a third determination in response to the first determination and the second determination to determine that the length of time is greater than a predetermined period of time; as well as The automated driving vehicle is operated to accelerate or decelerate in response to the third determination.

37. The method of claim 35, wherein said performing said second determination comprises: In response to determining from the image that a width of the lane is within a range of a predefined standard width for a standard lane, and in response to determining that a trajectory for the automated driving vehicle to turn away from a center of the lane to a side of the lane is available, determining that the risk level is low, Wherein said predetermined time period is associated with said risk level being low.

38. The method of claim 35, wherein said performing said second determination comprises: The risk level is determined to be medium in response to the following circumstances: Determining, based on the image, that the width of the lane is smaller than a predefined standard width of a standard lane, or determining that a trajectory for the automated driving vehicle to turn away from the center of the lane to a side of the lane is not available, or determining that the lane includes a curved portion; Wherein said predetermined time period is associated with said risk level being medium.

39. The method according to claim 35, The performing of the second determination comprises: determining that the risk level is high in response to determining from the set of images that the autonomous driving vehicle is parallel to or within a distance parallel to the vehicle that is turning; and The method further includes, in response to the determination, operating the autonomous driving vehicle to accelerate, decelerate, or change lanes.

40. The method of claim 35, further comprising: determining, based on at least one image in the image set, that a length of the vehicle is greater than a predetermined length; as well as In response to the determination, the autonomous driving vehicle is operated to steer away from the center of the lane to a side of the lane.

41. A system for operating an autonomous vehicle, comprising a computer including a processor configured to perform the method according to any one or more of claims 1 to 39.

42. A non-transitory computer readable program storage medium having stored thereon codes which, when executed by a processor, cause the processor to perform the method according to any one or more of claims 1 to 39.

Images