Vehicle systems and procedures for autonomous operation with unclassified hazard detection

Abstract

Systems and procedures for detecting unclassified hazards are provided for a vehicle's driver assistance system.The systems comprise a data storage device containing map data, one or more sensors configured to generate sensor data indicating one or more detected objects near the vehicle, and a controller configured to receive the map data and sensor data by means of one or more processors, to filter the sensor data to omit at least one of the detected objects located outside a boundary area that includes part of an adjacent lane next to a current lane in which the vehicle is operating, to determine whether an unclassified hazard is present in the adjacent lane, and to maintain the operation of the vehicle in the current lane in response to a determination that the unclassified hazard is present in the adjacent lane.

Description

IntroductionThe technical field relates generally to vehicle systems, and more particularly to autonomous operation of a vehicle configured to detect unclassified hazards to promote safety during automatic lane change.An autonomous vehicle is a vehicle that is capable of detecting its environment and navigating with little or no user input. An autonomous vehicle captures its environment using sensing devices such as radar, lidar, image sensors, and the like. The autonomous vehicle system also utilizes information from global positioning systems (GPS) technology, navigation systems, vehicle-to-vehicle communication, vehicle-to-infrastructure technology, and / or drive-by-wire systems to navigate the vehicle.Vehicle automation has been classified into numerical levels ranging from zero, i.e., no automation with full human control, to five, i.e., full automation without human control. Various automated driver assistance systems such as cruise control, adaptive cruise control and parking assistance correspond to lower degrees of automation, while genuine "sensorless" vehicles correspond to higher degrees of automation.Due to the plurality of different variables in a real-world environment, an autonomous vehicle control system may encounter an environment or scenario in which assistance is required. For example, traffic, road conditions, and other obstacles or scenarios may occur that affect autonomous operation. In lower level automation systems (e.g., level 3 or below), in such cases, a driver or other vehicle occupant may be required to manually control or operate the vehicle, which represents a burden on the driver that is counter to the purpose of automation.DE 10 2014 212 478 A1 describes a method for creating a surroundings model of a vehicle, wherein a gate is determined based on objects and / or free space boundaries and / or roadway boundaries, wherein the gate specifies the freely drivable region around the vehicle, wherein the gate comprises at least one gate segment, wherein the gate segment comprises at least one gate segment boundary, in particular a front gate segment boundary, a rear gate segment boundary, a left gate segment boundary and a right gate segment boundary, wherein the distance related to the gate segment boundary is determined, and wherein the gate is made available to a driver assistance system.DE 10 2014 212 704 B4 describes an adaptive speed control / control system which controls / controls the speed of the own vehicle by taking into account the predicted behavior of a preceding vehicle which travels in front of the own vehicle in the same lane or in an adjacent lane, and in particular an adaptive speed control / control system which also takes into account the predicted behavior of the own vehicle.DE 10 2019, 217 006 A1 describes a method, a device and a computer program product for determining load-minimal trajectories for the route planning of vehicles, in particular vehicles driving autonomously. In this case, along a road section on a local level, the loads acting on a vehicle and its occupants are detected and evaluated. As a result, the life of the vehicle is extended and the risk of chinetosis for the occupants is significantly reduced. Following vehicles can use the results to optimize their trajectories to be traveled on. In order to achieve this, trajectories already traveled and / or known load hot spots are called up when traveling a road section, and a trajectory is sought on the basis of this trajectory in which the loads for the vehicle and occupant can be further reduced or avoided. While this stress minimal trajectory is being traveled on, it is detected along this trajectory whether stress hot spots occur there.DE 10 2020 200 183 A1 describes a method for creating a probabilistic free space map with static and dynamic objects, with the following steps: retrieving static objects and a perception range polygon from an existing environment model; collecting predicted trajectories of dynamic objects; joining the static objects, the characteristics of determining the prediction of the perceived range polygon and the predicted trajectories in a first free space map; defining a maximum prediction time; defining prediction time steps; defining a current prediction time and setting this current prediction time to the value 0 for defining the start of a defined prediction time period; defining confidence ranges around the static and dynamic objects; defining at least one nonsecure range around at least one static or dynamic object; generating a first probabilistic free space map for the current prediction time; generating at least one further free space map for at least one prediction time step; evaluating the generated free space maps.Accordingly, it is desirable to provide vehicle control systems and methods capable of autonomously solving scenarios and finding a satisfactory solution for autonomous operation of the vehicle with reduced driver burden to improve the user experience. Other desirable features and characteristics of the present invention will become apparent from the following detailed description and the appended claims, taken in conjunction with the accompanying drawings and the foregoing technical field and background.DESCRIPTION OF THE INVENTIONA method for detecting and avoiding unclassified hazards is provided to a driver assistance system of a vehicle. In one embodiment, the method includes obtaining map data from a data storage device including information regarding a road including lane geometry, obtaining sensor data from one or more sensors of the vehicle including one or more detected objects proximate the vehicle, filtering the sensor data with the one or more processors to omit at least one of the one or more detected objects that is outside a boundary region including a portion of an adjacent lane adjacent to a current lane on which the vehicle is operating, the boundary region having a size determined at least in part using the map data and a current position of the vehicle, the one or more processors determining whether an unclassified danger is present in the adjacent lane, based at least in part on an amount of the one or more detected objects located within the bounding region, and wherein the one or more processors maintain operation of the vehicle in the current lane in response to a determination that the unclassified danger is present in the adjacent lane.In various embodiments, the method includes preventing, by the driver assistance system having the one or more processors, autonomous lane change in response to a determination that the unclassified risk is present in the adjacent lane.In various embodiments, the method includes generating, by the one or more processors, a normalized bounding area that includes an area that is less than an area of the bounding area based on a normalized error of the one or more sensors.In various embodiments, the method includes, by the one or more processors: determining a confidence score based at least in part on the amount of the one or more detected objects located within the boundary region, determining a lane score based on the confidence score, and comparing the lane score to a lane score threshold, wherein operation of the vehicle in the current lane is maintained in response to the lane score being greater than the lane score threshold.In various embodiments, the method includes determining the confidence value based on at least one of the following factors: time of detection of the one or more detected objects, object-to-object position correlation of the one or more detected objects, number of the one or more detected objects located within the boundary region, and a pattern defined by the one or more detected objects.In various embodiments, the method includes obtaining vision data from one or more cameras onboard the vehicle that include the portion of the adjacent lane, wherein the size of the bounding area is based at least in part on the map data, the current position of the vehicle, and the vision data.In various embodiments, the method includes activating, by the one or more processors, an autonomous lane change in response to a determination that the unclassified risk is absent in the adjacent lane.In various embodiments, the method includes dynamically adjusting a shape of the boundary region to correlate with the geometry of the adjacent lane.In various embodiments, the one or more sensors includes a radar sensor and the sensor data includes radar point cloud data.A system for detecting and avoiding unclassified hazards is provided to a vehicle driver assistance system. In one embodiment, the system includes a data storage device having map data containing information regarding a road including lane geometry, one or more sensors of the vehicle configured to generate sensor data indicative of one or more sensed objects proximate the vehicle, and a controller in operable communication with the data storage device and the one or more sensors. The controller is configured to, by one or more of its processors, receive the map data and the sensor data, filter the sensor data, to omit at least one of the one or more detected objects located outside a boundary region that includes a portion of an adjacent lane adjacent to a current lane on which the vehicle is operating, the boundary region having a size determined at least in part by the map data and a current position of the vehicle, determine whether an unclassified risk is present in the adjacent lane based at least in part on a quantity of the one or more detected objects located within the boundary region, and maintain operation of the vehicle in the current lane in response to a determination that the unclassified risk is present in the adjacent lane.In various embodiments, the controller is configured to prevent, via the one or more processors, autonomous lane change via the driver assistance system in response to a determination that the unclassified risk is present in the adjacent lane.In various embodiments, the controller is configured to generate, via the one or more processors, a normalized bounding area comprising an area less than an area of the bounding area based on a normalized error of the one or more sensors.In various embodiments, the controller is configured to, by the one or more processors: determine a confidence score that determines a lane score based at least in part on the amount of the one or more detected objects located within the boundary region, based on the confidence score, and compare the lane score to a lane score threshold, wherein operation of the vehicle in the current lane is maintained in response to the lane score being greater than the lane score threshold.In various embodiments, the controller is configured to determine, by the one or more processors, the confidence value based on at least one of the following factors: time of detection of the one or more detected objects, object-to-object position correlation of the one or more detected objects, amount of the one or more detected objects located within the boundary region, and a pattern defined by the one or more detected objects.In various embodiments, the controller is configured to, by the one or more processors: obtain, from one or more cameras onboard the vehicle, vision data including the portion of the adjacent lane, and wherein the size of the boundary region is based at least in part on the map data, the current position of the vehicle, and the vision data.In various embodiments, the controller is configured to enable the autonomous lane change via the one or more processors in response to a determination that the unclassified risk is not present in the adjacent lane.In various embodiments, the controller is configured to dynamically adjust a shape of the bounding area by way of the one or more processors to correlate it with the geometry of the adjacent lane.In various embodiments, the one or more sensors includes a radar sensor and the sensor data includes radar point cloud data.A vehicle is provided that includes a data storage device including map data including information related to a road including lane geometry, one or more radar sensors of the vehicle configured to generate radar point cloud data including one or more radar objects proximate the vehicle, and a controller in operable communication with the data storage device and the one or more radar sensors. The controller is configured to, by one or more of its processors: receive the map data and the radar point cloud data, filter the radar point cloud data, omit at least one of the one or more radar objects that is outside a boundary region that includes a portion of an adjacent lane adjacent to a current lane on which the vehicle is operating, the boundary region having a size determined at least in part by the map data and a current position of the vehicle, the map data, and a current position of the vehicle, wherein a shape of the boundary region is dynamically adjusted to correlate with the lane geometry, generate a normalized boundary region that includes a region that is less than a region of the boundary region based on a normalized error of the one or more radar sensors, determining a confidence value based on at least one of a detection time of the one or more radar objects, an object-to-object position correlation of the one or more radar objects, an amount of the one or more radar objects located within the boundary region, and / or a pattern defined by the one or more radar objects, determining a lane score based on the confidence value, comparing the lane score to a lane score threshold, determining whether an unclassified risk exists in the adjacent lane based on the lane score, wherein the unclassified risk is determined in response to the lane score being greater than the lane score threshold, operating the vehicle based on the determination, whether the unclassified danger is present in the adjacent lane, including: prohibiting autonomous lane change in response to determining that the unclassified danger is present in the adjacent lane; and enabling autonomous lane change in response to determining that the unclassified danger is not present in the adjacent lane.In various embodiments, the controller of the vehicle is configured to obtain, by the one or more processors: vision data comprising the portion of the adjacent lane from one or more cameras onboard the vehicle, wherein the size of the boundary region is based at least in part on the map data, the current position of the vehicle, and the vision data.Brief Description of the DrawingsThe exemplary embodiments are described below in connection with the following drawings, wherein like numerals designate like elements, and wherein: FIG. 1 is a functional block diagram illustrating an autonomous vehicle control system for a vehicle in accordance with various implementations; FIG. 2 is a block diagram of an automated driving system (ADS) suitable for implementation by the autonomous vehicle control system of the vehicle of FIG. 1, in accordance with various implementations; FIG. 3 is a dataflow diagram of an autonomous vehicle control system including an unclassified hazard detection system suitable for use with the ADS of FIG. 2 in the autonomous vehicle control system of FIG. 1, in accordance with one or more aspects described herein; FIG. 4 shows a flowchart of an unclassified hazard detection process suitable for implementation by the ADS of FIG. 2 in the autonomous vehicle control system of FIG. 1, in accordance with one or more aspects described herein; and FIGS. 5-10 show example scenarios for example implementations of the unclassified hazard detection process of FIG. 4, in accordance with one or more aspects described herein. FIG. 11 includes plots illustrating an example relationship between a confidence level and a corresponding lane score as determined by the unclassified hazard detection process of FIG. 4, in accordance with one or more aspects described herein. FIGS. 12 and 13 show example scenarios for an example implementation of the unclassified hazard detection process of FIG. 4, in accordance with one or more aspects described herein.DETAILED DESCRIPTIONThe following detailed description is merely exemplary in nature and is not intended to limit the application and use. Moreover, there is no intention to be bound by any explicit or implied theory set forth in the preceding technical field, background, brief summary, or the following detailed description. As used herein, the term module refers to any hardware, software, firmware, electronic control component, processing logic, and / or processor device, alone or in any combination, including, but not limited to: application specific integrated circuits (ASIC), an electronic circuit, a processor (shared, dedicated, or group), and memory that executes one or more software or firmware programs, a combinational logic circuit, and / or other suitable components that provide the described functionality.Embodiments of the present disclosure may be described herein in terms of functional and / or logical block components and various processing steps. Such block components may be implemented by any number of hardware, software, and / or firmware components configured to perform the specified functions. For example, an embodiment of the present disclosure may use various integrated circuit components, for example, memory elements, digital signal processing elements, logic elements, look-up tables, or the like, which may perform a variety of functions under the control of one or more microprocessors or other controllers. Moreover, those skilled in the art will appreciate that embodiments of the present disclosure may be used in connection with any number of systems and that the systems described herein are merely exemplary embodiments of the present disclosure.For brevity, conventional techniques associated with signal processing, data transmission, signaling, control, and other functional aspects of the systems (and the individual operating components of the systems) will not be described in detail herein. Moreover, the connection lines depicted in the various figures are intended to represent example functional relationships and / or physical couplings between the various elements. It should be appreciated that many alternative or additional functional relationships or physical connections may be present in an embodiment of the present disclosure.Referring to FIG. 1, an autonomous vehicle control system 100, according to one or more implementations, determines a plan for autonomous operation of a vehicle 10 along a route in a manner that takes into account unclassified objects or obstacles detected by onboard sensors 28, 40, as described in more detail below. In this regard, a control module on-board the vehicle 10 calibrates various types of on-board sensors 28, 40 with respect to each other and / or to the vehicle 10, whereby data from these various types of on-board sensors 28, 40 may be spatially or otherwise associated with each other based on the calibration for purposes of object detection, object classification, and the resulting autonomous operation of the vehicle 10. In various implementations, the vehicle control system 100 includes an unclassified threat detection system configured to detect unclassified threats by at least partially correlating map data to radar point cloud data to promote safe autonomous lane change maneuvers.In various embodiments, the vehicle 10 may be any of a number of different types of vehicles, such as a sedan, wagon, truck, or sport utility vehicle (SUV), and may have two-wheel drive (2WD) (i.e., rear-wheel drive or front-wheel drive), four-wheel drive (4WD), or all-wheel drive (AWD), and / or various other types of vehicles in certain embodiments. In certain embodiments, the vehicle 10 may also include a motorcycle and / or one or more other types of vehicles. Moreover, in various embodiments, the vehicle 10 may also include any number of other types of mobile platforms.As shown in FIG. 1, the example vehicle 10 generally includes a chassis 12, a body 14, front wheels 16, and rear wheels 18. the body 14 is disposed on the chassis 12 and substantially encloses components of the vehicle 10. The wheels 16- 18 are each rotatably connected to the chassis 12 near a corner of the body 14.In example implementations, the vehicle 10 is an autonomous vehicle or otherwise configured to support one or more autonomous modes of operation, and the control system 100 is integrated with the vehicle 10. In an exemplary implementation, the vehicle 10 is a so-called level two automation system. A level two system refers to "partial automation of driving" and refers to driving mode-dependent performance of an automated driving system for controlling steering, acceleration, and braking in certain scenarios while a driver remains attentive and actively monitors the automated driving system at all times and is capable of assisting the driver in controlling primary driving tasks.As shown, the vehicle 10 generally includes a propulsion system 20, a transmission system 22, a steering system 24, a braking system 26, a sensor system 28, an actuator system 30, at least one data storage device 32, at least one controller 34, and a communication system 36. The transmission system 22 is configured to transmit power from the propulsion system 20 to the wheels 16, 18 according to selectable speed ratios. According to various embodiments, the transmission system 22 may include a continuously variable transmission, a continuously variable transmission, or another suitable transmission. The brake system 26 is configured to apply a braking torque to the vehicle wheels 16, 18. The brake system 26 may include friction brakes, cable brakes, a regenerative braking system such as an electric machine, and / or other suitable braking systems, in various embodiments. The steering system 24 affects the position of the wheels 16- 18. Although a steering wheel is shown for illustrative purposes, in some embodiments contemplated within the scope of the present disclosure, the steering system 24 may not include the steering wheel 24 a.The sensor system 28 includes one or more sensing devices 40 a- 40 nthat sense observable conditions of the exterior environment and / or the interior environment of the vehicle 10. The sensor devices 40 a- 40 nmay include, but are not limited to, radars, lidars, global positioning systems, optical cameras, thermal cameras, ultrasonic sensors, and / or other sensors. The actuator system 30 includes one or more actuator devices 42 a- 42 nthat control one or more vehicle functions, such as, but not limited to, the propulsion system 20, the transmission system 22, the steering system 24, and the brake system 26. In various embodiments, the vehicle features may also include interior and / or exterior features of the vehicle, such as doors, a trunk, and cabin features such as air, music, lighting, etc. (not numbered).The data storage 32 stores data used for automatic control of the vehicle 10. In various implementations, the data storage device 32 stores defined navigable environment maps. In various embodiments, the defined maps may be predefined by and obtained from a remote system. For example, the defined maps may be compiled by the remote system and transmitted to the vehicle 10 (wirelessly and / or wired) and stored in the data storage device 32. As can be appreciated, the data storage device 32 may be part of the controller 34, separate from the controller 34, or part of the controller 34 and part of a separate system. The storage device 32 may be any suitable type of storage device, including various types of random access memories and / or other storage devices. In an exemplary embodiment, the storage device 32 includes a program product from which a computer readable storage device may receive a program that executes one or more embodiments of one or more processes of the present disclosure, such as the steps of the process discussed further below in connection with FIG. 4. In another example embodiment, the program product may be directly stored in and / or otherwise accessed by the storage device and / or one or more other disks and / or other storage devices.The controller 34 includes at least a processor 44, a communication bus 45, and a computer readable storage device or medium 46. the processor 44 performs the computation and control functions of the controller 34. Processor 44 may be any custom or commercially available processor, a central processing unit (CPU), a graphics processing unit (GPU), an auxiliary processor among multiple processors associated with controller 34, a semiconductor-based microprocessor (in the form of a microchip or chipset), a macroprocessor, any combination thereof, or generally any device for executing instructions. The computer readable storage devices or media 46 may include volatile and nonvolatile memories, for example, read only memory (ROM), random access memory (RAM), and keep alive memory (KAM). KAM is a persistent or non-volatile memory that can be used to store various operating variables while the processor 44 is off. The computer readable storage device(s) 46 may / may be implemented using any number of known storage devices, such as programmable read only memory (PROMs), programmable read only memory (EPROMs), electrically erasable PROM (EEPROMs), flash memory, or other electrical, magnetic, optical, or combined storage devices capable of storing data, some of which represent executable instructions, used by the controller 34 in controlling the vehicle 10. Bus 45 is used to transfer programs, data, status, and other information or signals between the various components of vehicle 10. These include, but are not limited to, direct hardwired connections, fiber optic technology, infrared and wireless bus technologies.The instructions may include one or more separate programs, each of which includes an ordered listing of executable instructions for implementing logical functions. When executed by the processor 44, the instructions receive and process signals from the sensor system 28, perform logic, calculations, methods, and / or algorithms for automatically controlling the components of the vehicle 10, and generate control signals for the actuator system 30 to automatically control the components of the vehicle 10 based on the logic, calculations, methods, and / or algorithms. Although only one controller 34 is shown in FIG. 1, embodiments of the vehicle 10 may include any number of controllers 34 that communicate and cooperate via any suitable communication medium or combination of communication media to process the sensor signals, perform logic, calculations, methods, and / or algorithms, and generate control signals to automatically control features of the vehicle 10.It will be appreciated that the controller 34 may vary from the embodiment shown in FIG. 1. For example, the controller 34 may be coupled to or otherwise utilize one or more remote computing systems and / or other control systems, for example, as part of one or more of the above-mentioned vehicle devices and systems. While this exemplary embodiment will be described in the context of a fully functional computer system, those skilled in the art will appreciate that the mechanisms of the present disclosure may be distributed as a program product with one or more types of non-transitory computer readable signal bearing media used to store the program and its instructions and perform its distribution, such as a non-transitory computer readable medium bearing the program and including computer instructions stored therein for causing a computer processor (such as processor 44) to perform and execute the program. Such a program product may take a variety of forms, and the present disclosure applies equally regardless of the particular type of computer readable signal bearing medium used to perform the distribution. Examples of signal-bearing media are writable media such as floppy disks, hard disks, memory cards and optical data carriers and transmission media such as digital and analog communication links. In certain embodiments, cloud-based storage and / or other techniques may also be used. It will also be appreciated that the computer system of the controller 34 may also otherwise differ from the embodiment shown in FIG. 1, for example, in that the computer system of the controller 34 may be coupled to or otherwise utilize one or more remote computer systems and / or other control systems.In various embodiments, one or more instructions of the controller 34 are included in the control system 100 (e.g., the data storage device 46) and, when executed by the processor 44, cause the processor 44 to obtain data captured or generated by imaging and ranging devices (e.g., the sensors 28, 40) and use the captured environmental data to determine commands for autonomous operation of the vehicle 10, as described in more detail below.In example implementations further related to FIG. 1, the communication system 36 is configured to wirelessly communicate information to and from other entities 48 via a communication network, such as, but not limited to, other vehicles ("V2V" communication), infrastructure ("V2I" communication), remote systems, and / or personal devices. In an example implementation, communication system 36 is a wireless communication system configured to communicate over a wireless local area network (WLAN) using IEEE 802.11 standards or using cellular data communication. However, additional or alternative communication methods, such as a dedicated short-range communication channel (DSRC), are also contemplated within the scope of the present disclosure. DSRC channels refer to one or two-sided short to medium range wireless communication channels that have been specially developed for use in automobiles, as well as a number of protocols and standards.The communication network utilized by communication system 36 may comprise a wireless carrier system, such as a mobile telephone system, that includes a plurality of cellular towers (not shown), one or more mobile switching centers (MSCs) (not shown), as well as any other networking components required to connect the wireless carrier system to a land communication system, and the wireless carrier system may implement any suitable communication technology, including, for example, digital technologies such as CDMA (e.g., CDMA2000), LTE (e.g., 4G LTE or 5G LTE), GSM / GPRS, or other current or emerging wireless technologies. Additionally or alternatively, a second wireless carrier system in the form of a satellite communication system may be used to provide unidirectional or bidirectional communication using one or more communication satellites (not shown) and an uplink transmitting station (not shown), including, but not limited to, satellite radio services, satellite telephony services, and / or the like. Some implementations may utilize a land-based communication system, such as a conventional land-based telecommunications network, including a public switched telephone network (PSTN) used to provide hardwired telephony, packet-switched data communication, and the Internet infrastructure. One or more segments of a land-based communication system may be implemented via a standard wired network, a fiber or other optical network, a cable network, power lines, other wireless networks such as wireless local area networks (WLANs) or broadband wireless access networks (BWA), or any combination thereof.As shown in FIG. 2, in accordance with various implementations, the controller 34 implements an autonomous driving system (ADS) 70. that is, suitable software and / or hardware components of the controller 34 (e.g., processor 44 and computer readable storage device 46) are used to provide an autonomous driving system 70 that is used in conjunction with the vehicle 10 to automatically control, for example, various actuators 30 and thereby control vehicle acceleration, steering, and braking, respectively, without human intervention.In various implementations, the instructions of the autonomous driving system 70 may be organized according to functions or systems. As shown in FIG. 2, the autonomous driving system 70 may include, for example, a sensor fusion system 74, a positioning system 76, a steering system 78, and a vehicle control system 80. As can be appreciated, in various implementations, the instructions may be organized (e.g., combined, further partitioned, etc.) into any number of systems, as the disclosure is not limited to the present examples.In various implementations, the sensor fusion system 74 synthesizes and processes sensor data and predicts the presence, location, classification, and / or path of objects and features of the environment of the vehicle 10. In various implementations, the sensor fusion system 74 may incorporate information from multiple sensors, including, but not limited to, cameras, lidars, radars, and / or any number of other types of sensors.The positioning system 76 processes sensor data along with other data to determine a position (e.g., a local position with respect to a map, an accurate position with respect to the lane of a road, the vehicle direction, the speed, etc.) of the vehicle 10 with respect to the environment. The steering system 78 processes sensor data along with other data to determine a path that the vehicle 10 is to follow given the current sensor data and the vehicle position. The vehicle control system 80 then generates control signals for controlling the vehicle 10 according to the determined path. In various implementations, the controller 34 implements machine learning techniques to support the functionality of the controller 34, for example, feature detection / classification, obstacle avoidance, route traversal, mapping, sensor integration, ground truth determination, and the like.In one or more embodiments, the steering system 78 includes a motion planning module that generates a motion plan for controlling the vehicle 10 as it travels along a route. The motion planning module includes a longitudinal release module that generates a longitudinal motion plan output for controlling the motion of the vehicle 10 along the route in the general direction of travel, for example, by causing the vehicle 10 to accelerate or decelerate at one or more locations in the future along the route to maintain a desired speed. The motion planning module also includes a lateral motion planning module that generates a lateral motion plan output for controlling the lateral motion of the vehicle 10 along the route to change the general direction of travel, for example, by steering the vehicle 10 at one or more locations in the future along the route (for example, to keep the vehicle 10 within the center of a lane, change lanes, etc.). The longitudinal and transverse plan outputs correspond to the commanded (or planned) path output provided to the vehicle control system 80 for controlling the vehicle actuators 30 to achieve movement of the vehicle 10 along the route corresponding to the longitudinal and transverse plans.In normal operation, the longitudinal release module attempts to optimize vehicle speed (or velocity) in the direction of travel, vehicle acceleration in the direction of travel, and the derivative of vehicle acceleration in the direction of travel, alternatively referred to herein as longitudinal jerk of the vehicle 10, and the lateral release module attempts to optimize one or more steering angles, the rate of change of the steering angle, and the acceleration or second derivative of the steering angle, alternatively referred to herein as lateral jerk of the vehicle 10. In this context, the steering angle may be related to the curvature of the path or the route, and the steering angle, the rate of change of the steering angle, and the acceleration or the second derivative of the steering angle may be optimized individually or in combination by the lateral solver module.In example implementations, the steering system 78 supports a hands-free, autonomous mode of operation that controls steering, acceleration, and braking while being activated and operating to center the lane while attempting to maintain a driver-selected speed and / or a driver-selected following distance (or gap time) relative to other vehicles using the current sensor data (or obstacle data) provided by the sensor fusion system 74 and the current vehicle position provided by the positioning system 76. In the autonomous mode of operation, the steering system 78 includes or otherwise implements a lane change coordinator that, in addition to data or other information from the sensor fusion system 74, the positioning system 76, and possibly other modules or systems, analyzes route information (if available) to determine whether or not to initiate and execute a lane change from a current lane to an adjacent lane, for example, based on the presence of slow-driving traffic within the current lane in front of the vehicle 10 (e.g., (e.g., to pass or pass another vehicle), whether the current lane ends or transitions to an adjacent lane, whether or not a lane change is required to maintain travel on the desired route, and / or the like. In this regard, the lane change coordinator may automatically determine when to initiate a lane change and automatically configure the lateral solution module and / or the motion planning module to generate a corresponding lateral plan to change the lane in the desired manner and provide the lateral plan to the vehicle control system 80 that automatically generates corresponding control signals for autonomously controlling the vehicle actuators 30 to maneuver the vehicle 10 and perform the lane change.Referring to FIG. 3, and with continued reference to FIGS. 1-2, a dataflow diagram illustrates elements of an unclassified hazard detection system 200 of the ADS 70 of FIG. 2 as implemented by the controller 34 of FIG. 1, in accordance with various embodiments. As can be appreciated, various embodiments of the unclassified hazard detection system 200 according to the present disclosure may include any number of modules embedded within the controller 34 that may be combined and / or further partitioned to implement the systems and methods described herein in a similar manner. Additionally, inputs to the unclassified hazard detection system 200 may be received from other control modules (not shown) connected to the vehicle 10 and / or determined / modeled by other sub-modules (not shown) within the controller 34. Moreover, the inputs may also be subjected to preprocessing such as subsampling, noise reduction, normalization, feature extraction, missing data reduction, and the like. In various embodiments, the unclassified hazard detection system 200 includes a localization module 210, a boundary region module 212, a hazard detection module 214, and a control module 216.In various embodiments, the localization module 210 receives as input map data 220 retrieved from the data storage device 32 or the computer readable storage device or media 46. The map data 220 includes various data indicating various characteristics of the road including road geometry and road curvature. In various embodiments, the localization module 210 receives as input sensor data 222 generated by the sensor system 28. The sensor data 222 includes various data indicative of camera objects including lane markings and radar objects.The localization module 210 performs localization based on a known road geometry. In various embodiments, the localization may include determining a location of the vehicle 10 based on a GPS signal. In various embodiments, the localization may include determining a lane of the road on which the vehicle 10 is located based on the lane markings. In various embodiments, the localization may include determining an approximate width of the lanes of the road based on the lane markings. The localization module 210 generates localization data 224 that includes various data indicative of, for example, the position of the vehicle 10, the lane in which the vehicle 10 is located, the approximate width of the lanes of the road, and / or the positions of the radar objects located on the road (i.e., radar objects located outside the road may be eliminated or omitted).In various embodiments, the bounding box module 212 receives as input the localization data 224 generated by the localization module 210. The bounding area module 212 determines an area of interest in a lane of the road, for example, a lane laterally adjacent to the lane in which the vehicle 10 is located, and defines a bounding box or area whose boundaries are associated with the lateral extents of the area of interest. In various embodiments, the region of interest corresponds to a portion of the road representative of portions of the road that would be physically entered during an autonomous lane change of the vehicle 10, wherein the presence of a persistent object (e.g., a concrete barrier, a crack in the road, etc.) within that portion of the road may pose a risk to the vehicle 10 when the autonomous lane change is performed. In various embodiments, the size or lateral dimension (e.g., width) of the boundary region may be determined at least in part by a predetermined standard deviation of the noise attributed to a radar system of the vehicle 10 that generated the radar objects of the sensor data 222. In various embodiments, the bounding area may have a dynamic shape that is continuously adjusted based at least in part on the curvature of the road determined based on the map data 220. The bounding area module 212 eliminates or omits at least one, some, or all of the radar objects that are outside the boundaries of the bounding area. In various embodiments, the bounding area may be normalized one or more times to obtain narrower bounding areas and create an adjusted distribution of the radar objects. In various embodiments, the boundary region may be normalized to obtain a normalized line in response to a repeating pattern of radar objects in laterally similar positions along an extended longitudinal region, which may represent a lateral position of a inferred threat (e.g., FIG. 9 ). The bounding area module 212 generates bounding area data 226 that includes various data indicating the specified bounding area as well as the amount and position of the radar objects within the bounding area.In various embodiments, the threat detection module 214 receives as input the bounding area data 226 generated by the bounding area module 212. The risk detection module 214 dynamically determines a confidence level associated with the likelihood that there is a risk within the bounding area based at least in part on the set of radar objects located within the bounding area at a particular time. In various embodiments, the confidence level may be expressed as a numerical value or percentage (e.g., 0 to 100 percent confidence that there is a risk). The hazard detection module 214 then dynamically generates and updates a lane score for the desired lane based at least in part on the confidence level. In various embodiments, the lane score may be expressed as a numerical value (e.g., between 100 and -100). In various embodiments, the lane score may be expressed as a word, sentence, and / or category (e.g., between "highly desirable" and "highly undesirable"). In various embodiments, the lane score may be determined as a numerical value and converted into a word, sentence, and / or category. For example, if the confidence level is 100 percent (i.e., very high or substantially safe), the lane score may be -100 (very low), which may be converted to "very undesirable.". In various embodiments, a relationship between the confidence level and the lane score may be weighted and / or exponential. The hazard detection module 214 generates hazard detection data 228 including various data indicative of the lane rating for one or more lanes of the road, such as the desired lane.In various embodiments, the control module 216 receives as input the threat detection data 128 generated by the threat detection module 214. The control module 216 generates instructions configured to permit and / or restrict lane change maneuvers based on the lane score. In various embodiments, the instructions are configured to enable and / or disable an autonomous lane change. The control module 216 generates control data 230 that includes various data indicative of the instructions. The control module 216 may transmit the control data 230 to the vehicle control system 80. In some embodiments, the control data 230 may cause the vehicle control system 80 and a lane change system thereof to operate normally in response to a sufficiently high lane rating (i.e., a low risk of risk present), and may suspend, disable, or prevent automatic lane change by the lane change system in response to a sufficiently low lane rating (i.e., a high risk of risk present), at least in a direction of the inferred risk. For example, in response to a sufficiently low lane rating, the control data 230 may cause operation of the vehicle 10 to be maintained in the current lane in which the vehicle 10 is operated until the lane rating increases sufficiently that autonomous lane change may be safely performed. In particular, in some embodiments, preventing the lane change to the adjacent lane having a low lane rating may not prevent the vehicle 10 from autonomously changing the lane to another lane having a sufficiently high lane rating.In various embodiments, the unclassified hazard detection system 200 may be continuously operating during operation of the vehicle 10, during operation of the lane change system, or in response to a request for autonomous lane change by the lane change system.FIG. 4 shows an example implementation of a method for detecting unclassified hazards 300 suitable for implementation by one or more control modules onboard a vehicle (e.g., the unclassified hazard detection system 200 in conjunction with the ADS 70 supported by the controller 34 in the vehicle 10) to autonomously operate one or more actuators onboard the vehicle to change lanes in a manner affected by the detection of unclassified hazards in lateral lanes. For purposes of illustration, the following description may refer to elements mentioned above in connection with FIGS. 1-3. While portions of the unclassified threat detection method may be performed by various elements of a vehicle system, for purposes of explanation, the subject matter may be described herein primarily in connection with the unclassified threat detection method performed primarily by the unclassified threat detection system 200 and the ADS 70 implemented by the controller 34 connected to the vehicle 10.To illustrate various aspects of the method 300, reference is made to the examples illustrated in FIGS. 5-13. In these examples, a vehicle 510 is operating on a road 500 that includes five lanes, including, from right to left, a first lane 501, a second lane 502, a third lane 503, a fourth lane 504, and a fifth lane 505. In FIGS. 5-7 and 12, the vehicle 510 is in the third lane 503, and in FIGS. 8-10 and 13, the vehicle 510 is in the first lane 501.In example implementations, the unclassified hazard detection method 300 may begin at 310. At 312, the method 300 may include receiving external object-related sensor data from one or more sensing devices 28, 40 of the vehicle 10, such as radar objects (i.e., targets / tails) from a radar system. In various embodiments, the radar objects may be stored as a cloud of radar points.At 314, the method 300 may include receiving map data from a data storage device (e.g., the data storage device 32 or the computer readable storage device or medium 46) that includes road features such as road geometry and road curvature. Optionally, at 314, the method 300 may include receiving road property-related sensor data from one or more sensing devices 28, 40 of the vehicle 10, such as video from a camera system. In various embodiments, the road property-related sensor data may be analyzed to determine a current lane of the road on which the vehicle 10 is located based on the lane markings. In various embodiments, the road property-related sensor data may be analyzed to determine an approximate width of the lanes of the road based on the lane markings. In various embodiments, the method 300 may include receiving lane markings, identifying a current lane being traveled by the vehicle 10, and / or lane widths determined by another system of the vehicle 10 at 314. In various embodiments, the method 300 may include determining a merged lane boundary based at least in part on a GPS position of the vehicle 10, map-based road features and curvatures, and camera-based perceived road boundaries.At 316, the method 300 may include setting a boundary region having boundaries associated with lateral extents of a region of interest in a lane of the road, such as a lane laterally adjacent to the current lane in which the vehicle 10 is located. In various embodiments, the size of the boundary region may be determined at least in part by a predetermined standard deviation of the noise associated with a radar system of the vehicle 10 that generated the radar objects. In various embodiments, the boundary region may have a dynamic shape that is continuously adjusted based at least in part on the road curvature.For example, in FIG. 5, a first bounding area defined by the boundaries 512a is set laterally to the left of the vehicle 510, and in FIG. 8, a second bounding area defined by the boundaries 512b is set laterally to the left of the vehicle 510. Specifically, the shapes of the first and second boundary regions have been adjusted based on the curvature of the road 500. For example, the second boundary region is curved to the right due to the curvature of the first lane 501 and the second lane 502 of the road 500. The radar objects 514 are displayed on the road 500 in the areas adjacent to the vehicle 510.At 318, the method 300 may include filtering the radar objects to eliminate or omit all radar objects that are outside the bounding region. At 320, the method 300 may include normalizing the boundary region one or more times to create a narrower boundary region and thereby generate an adjusted distribution of the radar objects. In various embodiments, one or more radar objects that are outside the normalized bounding range may be eliminated or omitted.In FIGS. 6 and 9, for example, the radar objects 514 of FIGS. 5 and 8 have been filtered such that only certain radar objects 516 remain which are located within the delimiting regions 512. In these examples, there are four remaining radar objects 516 awithin the first bounding range and four remaining radar objects 516 bwithin the second bounding range. In FIGS. 7 and 10, the bounding areas 512 are normalized based on the positions of the remaining radar objects 516 to obtain normalized bounding areas 520.At 322, method 300 may include determining a level of confidence associated with the likelihood that a risk exists within the bounding area based at least in part on the amount of radar objects located within the bounding area at a particular time. Depending on the set of radar objects within the boundary area and / or a grouping of the radar objects within the boundary area, the confidence level may be dynamically increased or decreased. For example, an increase in the number of radar objects within the normalized bounding range, a pattern of the radar objects within the normalized bounding range, and / or persistence (i.e., repetition) of radar objects within the normalized bounding range over time may result in an increase in the confidence level at 324. In contrast, a decrease in the number of radar objects within the normalized bounding region and / or a lack of persistence (i.e., repetition) of radar objects within the normalized bounding region over time may result in a decrease in the confidence level at 326. In various embodiments, the confidence level may increase exponentially. For example, the confidence level may increase with increasing speed as the radar objects persist within the normalized bounding region. In various embodiments, the confidence level may decrease or decay relatively slowly over time, for example in response to a decrease in persistence of the radar objects within the normalized boundary region, rather than in the form of a sharp decrease or a stepwise change.At 328, the method 300 may include determining or establishing a lane score based at least in part on the confidence level. In various embodiments, the lane score may be dynamically updated based on the confidence level over time. The lane score may have various confidence level relationships, such as, but not limited to, an exponential relationship. In various embodiments, the lane score may be a composition based on the level of confidence and one or more additional factors. In various embodiments, the lane score may increase exponentially. For example, the lane rating may increase with an increasing rate, for example, depending on the durability of the confidence level. In various embodiments, the lane score may decrease or decay at a predetermined rate over time, for example in response to a decrease in the durability of the confidence level, rather than decreasing with a sharp decrease or a stepwise change.For example, FIG. 11 shows a first (upper) graph illustrating a confidence level increasing over time and a second (lower) graph illustrating a lane score decreasing over time based on the confidence level of the first graph. The relationship between the confidence level of the first graph and the lane score of the second graph shown in FIG. 11 is not limiting. In the first plot, the y-axis is a percent confidence level (%) and the x-axis is time (t) in seconds. In the second graph, the y-axis is a unitless lane score and the x-axis is time (t) in seconds.At 330, the lane score may be compared to a pre-programmed lane score threshold. At 332, method 300 may include suspending or preventing a lane change maneuver in response to the lane score exceeding the lane score threshold. At 334, method 300 may include enabling or permitting a lane change maneuver in response to the lane score being below the lane score threshold. The method 300 may end at 336.In various embodiments, the boundary region may be normalized to generate a normalized line in response to a repeating pattern of radar objects in laterally similar positions along an extended longitudinal region, which may represent a lateral position of a suspected threat. For example, FIG. 12 illustrates a first normalized line 522 aadjacent to the vehicle 510, and FIG. 13 illustrates a second normalized line 522 badjacent to the vehicle 510.The systems and methods disclosed herein provide various advantages over certain existing systems and methods. For example, various autonomous driving systems perform object and / or hazard detection by using sensor inputs to classify an object. Although useful in many situations, classification of objects may require special purpose sensors and / or substantial computational resources. Problems may also occur when there are objects that do not fall within one of the preprogrammed classification categories. The systems and methods presented herein solve these problems by enabling recognition and avoidance of unclassified hazards by correlating map data and sensor data (e.g., radar objects) to recognize, determine, and / or avoid unclassified hazards on a road, and disabling and / or enabling one or more autonomous functions (e.g., automatic lane change maneuvers) to avoid such hazards. By combining implicit knowledge about the road included in the map data with sensor data from the vehicle, the systems and methods disclosed herein are capable of providing multi-domain, multi-sensor fusion interpretation based on, for example, a single sensor data (e.g., radar). In various embodiments, the systems and methods are not required to classify the hazards. For example, if the systems and methods enable or suspension automatic lane change maneuvers, knowing the probability of a potential unclassified risk in a lateral lane is sufficient to determine whether the vehicle should change from a safe position (i.e., the current lane) to a potentially unsafe position (e.g., the desired lane). By detecting potential hazards without classifying them, the systems and methods disclosed herein may promote safe operation of the vehicle while reducing the type or amount of sensors and / or computational requirements required to operate the ADS 70.Although at least one exemplary embodiment has been presented in the foregoing detailed description, it should be understood that there are a variety of variations. It should also be appreciated that the example embodiment or embodiments are only examples and are not intended to limit the scope, applicability, or configuration of the disclosure in any way. Rather, the foregoing detailed description is intended to teach a person of ordinary skill in the art a practical guide to the implementation of the example embodiment or embodiments. It is to be understood that various changes may be made in the function and arrangement of elements without departing from the scope of the disclosure as set forth in the appended claims and their legal equivalents.

Claims

A method for detecting and avoiding unclassified hazards to a driver assistance system of a vehicle (10), comprising: obtaining, from a data storage device (32), map data (220) containing information relating to a road including lane geometry; obtaining, from one or more sensors (28, 40) of the vehicle (10), sensor data (222) containing one or more detected objects in the vicinity of the vehicle (10); Filtering, with the one or more processors (44), the sensor data (222) to omit at least one of the one or more detected objects located outside a boundary region comprising a portion of an adjacent lane adjacent to a current lane on which the vehicle (10) is operating, the boundary region having a size determined at least in part using the map data (220) and a current position of the vehicle (10); determining, with the one or more processors (44), whether an unclassified risk is present in the adjacent lane based at least in part on an amount of the one or more detected objects located within the boundary region; and maintaining, with the one or more processors (44), operation of the vehicle (10) in the current lane in response to a determination that the unclassified risk is present in the adjacent lane.The method of claim 1, further comprising preventing, with the one or more processors (44), autonomous lane change via the driver assistance system in response to a determination that the unclassified risk is present in the adjacent lane.The method of claim 1, further comprising, by the one or more processors (44): determining a confidence score based at least in part on the amount of the one or more detected objects located within the boundary area; determining a lane score based on the confidence score; and comparing the lane score to a lane score threshold, wherein operation of the vehicle (10) in the current lane is maintained in response to the lane score being greater than the lane score threshold.The method of claim 1, further comprising dynamically adjusting a shape of the boundary region to correlate with the geometry of the adjacent lane.The method of claim 1, wherein the one or more sensors (28, 40) comprise a radar sensor and the sensor data (222) includes radar point cloud data.A system for detecting and avoiding unclassified hazards for a driver assistance system of a vehicle (10), comprising: a data storage device (32) comprising map data (220) containing information relating to a road including lane geometry; one or more sensors (28, 40) of the vehicle (10) configured to generate sensor data (222) indicative of one or more detected objects in the vicinity of the vehicle (10); and a controller (34) in operable communication with the data storage device (32) and the one or more sensors (28, 40), wherein the controller (34) is configured to receive, by means of one or more of its processors (44): the map data (220) and the sensor data (222); filtering the sensor data (222) to omit at least one of the one or more detected objects located outside a boundary region comprising a portion of an adjacent lane adjacent to a current lane on which the vehicle (10) is operating, the boundary region having a size determined at least in part using the map data (220) and a current position of the vehicle (10); determine whether an unclassified danger is present in the adjacent lane based at least in part on an amount of the one or more detected objects located within the boundary region; and maintain operation of the vehicle (10) in the current lane in response to a determination that the unclassified danger is present in the adjacent lane.The system of claim 6, wherein the controller (34) is configured to prevent, via the one or more processors (44), autonomous lane change via the driver assistance system in response to a determination that the unclassified risk is present in the adjacent lane.The system of claim 6, wherein the controller (34) is configured to, by the one or more processors (44): determine a confidence score based at least in part on the amount of the one or more detected objects located within the bounding area; determine a lane score based on the confidence score; and compare the lane score to a lane score threshold, wherein operation of the vehicle (10) in the current lane is maintained responsive to the lane score being greater than the lane score threshold.The system of claim 6, wherein the controller (34) is configured to dynamically adjust a shape of the bounding area by the one or more processors (44) to correlate with the geometry of the adjacent lane.The system of claim 6, wherein the one or more sensors (28, 40) comprise a radar sensor and the sensor data (222) includes radar point cloud data.

Images