VEHICLE WITH ENVIRONMENTAL CONTEXT ANALYSIS

Abstract

Carrier vehicle, including: a motor, brakes, sensors and a processor, configured to: to identify a lane of the target vehicle; to determine a radius of curvature of the road boundaries of the detected lane; to predict a target path of the target vehicle based on lane boundaries of a virtual map and the target path based on the specified radius of curvature; to compare the target path with a predicted carrier path of the carrier vehicle; and to apply the brakes based on the comparison.

Description

TECHNICAL AREA

[0001] This invention relates to an environmental context (e.g. lane markings) for vehicles. STATE OF THE ART

[0002] Existing vehicles are configured to predict a collision based on the speed of the existing vehicle and the speed of a target vehicle. For example, US 9 248 834 B1, US 2011 / 0 087 433 A1, and US 2013 / 0 099 911 A1 each demonstrate vehicles that detect target vehicles and avoid collisions with them. However, many of these predictions do not account for changes in the target vehicle's speed based on the surrounding context. For instance, if the target vehicle is traveling in a roundabout, its speed (where speed includes course or direction) will likely change to follow the roundabout. Thus, solutions are needed to incorporate the surrounding context as a factor in collision predictions. SUMMARY

[0003] The problem is solved by the features of the independent patent claims. Advantageous embodiments of the invention are described in the dependent claims.

[0004] A carrier vehicle may include: motor(s), brakes, sensors, processor(s), configured to: (a) predict a target path of a target vehicle based on lane boundaries of a virtual map; (b) compare the target path with a predicted carrier path of the carrier vehicle; (c) apply the brakes based on the comparison; (d) predict the target path and the carrier path as shapes, each having a two-dimensional area; (e) determine whether the shapes intersect, and based on the determination, (f) calculate a first time interval in which the target vehicle reaches the point of intersection, and (g) calculate a second time interval in which the carrier vehicle reaches the point of intersection. BRIEF DESCRIPTION OF THE DRAWINGS

[0005] For a better understanding of the invention, reference may be made to the embodiments shown in the following drawings. The components in the drawings are not necessarily to scale, and associated elements may be omitted, or in some cases, proportions may have been exaggerated to emphasize and clearly illustrate the novel features described herein. Furthermore, system components may be arranged in different ways, as is known in the prior art. Additionally, in the drawings, the same reference numerals in the various views denote corresponding parts. Fig. Figure 1 is a block diagram of a vehicle computer system. Fig. Figure 2 is a top view of a carrier vehicle that includes the vehicle's computer system. Fig. Figure 3 is a block diagram that corresponds to a collision hazard assessment. Fig. 4 is a more concrete embodiment of the block diagram of Fig. 3. Fig. 5 is a first traffic scenario and represents a kind of virtual map, presented in graphical form. Fig. Figure 6 shows possible operations associated with the first, second, and third traffic scenarios. Fig. 7 is a second traffic scenario and represents a type of virtual map presented in graphical form. Fig. Figure 8 illustrates the functions of a cross-traffic alert system. Fig. 9 is a third traffic scenario and represents a type of virtual map presented in graphical form. Fig. Figure 10 shows possible operations associated with the first, second and third traffic scenarios. Fig. Figure 11 shows possible operations associated with the first, second and third traffic scenarios. Fig. Figure 12 shows possible operations associated with the first, second and third traffic scenarios. Fig. Figure 13 shows possible operations associated with the first, second and third traffic scenarios. Fig. Figure 14 shows possible operations associated with the first, second and third traffic scenarios. DETAILED DESCRIPTION OF EXAMPLES OF EXECUTION

[0006] Although the invention may be embodied in various forms, some exemplary and non-limiting embodiments are shown in the drawings and described below, under the assumption that the present disclosure is to be regarded as an illustration of the invention and is not intended to limit the invention to the specific embodiments shown.

[0007] In this application, the use of an exclusive form should also include the inclusive meaning. The use of definite or indefinite articles should not indicate cardinality. In particular, reference to "the" object or "a" object should also denote one of a possible multitude of such objects. Furthermore, the conjunction "or" can be used to represent features that are present simultaneously, as one option, and mutually exclusive alternatives, as another option. In other words, the conjunction "or" should be understood as including "and / or" as one option and "either / or" as another option.

[0008] Fig. Figure 1 shows a computing system 100 of a carrier vehicle 200. The carrier vehicle 200 is connected, meaning it is configured to (a) wirelessly receive data from external units (e.g., infrastructure, servers, other connected vehicles) and (b) wirelessly transmit data to external units. The vehicle 200 can be autonomous, semi-autonomous, or manual. The carrier vehicle 200 includes a motor, a battery, at least one wheel driven by the motor, and a steering system configured to rotate the at least one wheel around an axle. The carrier vehicle 200 can be powered by fossil fuels (e.g., diesel, gasoline, natural gas), hybrid-electric, fully electric, fuel cell-powered, etc.

[0009] Vehicles are described, for example, in U.S. Patent Application No. 15 / 076,210 by Miller, U.S. Patent No. 8,180,547 by Prasad et al., U.S. Patent Application No. 15 / 186,850 by Lavoie et al., U.S. Patent Publication No. 2016 / 0117921 by D'Amato, and U.S. Patent Application No. 14 / 972,761 by Hu, all of which are hereby incorporated in full by reference. Vehicle 200 may include any of the features described in Miller, Prasad, Lavoie, D'Amato, and Hu.

[0010] The computer system 100 is integrated into the carrier vehicle 200. Among other things, the computer system 100 supports the automatic control of the carrier vehicle 200's mechanical systems and enables communication between the carrier vehicle 200 and external units (e.g., connected infrastructure, the internet, other connected vehicles). The computer system 100 includes a data bus 101, one or more processors 108, volatile memory 107, non-volatile memory 106, user interfaces 105, a telematics unit 104, actuators and motors 103, and local sensors 102.

[0011] The data bus 101 transmits electronic signals or data between the electronic components. The processor 108 performs operations on the electronic signals or data to generate modified electronic signals or data. The volatile memory 107 stores data for near-instantaneous retrieval by the processor 108. The non-volatile memory 106 stores data for retrieval by the volatile memory 107 and / or the processor 108. The non-volatile memory 106 includes a variety of non-volatile storage media, including hard disks, SSDs, DVDs, Blu-rays, etc. The user interface 105 includes displays, touchscreens, keyboards, buttons, and other devices that enable user interaction with the computing system. The telematics unit 104 enables both wired and wireless communication with external units via Bluetooth, cellular data (e.g., 3G, LTE), USB, etc.

[0012] The actuators / motors 103 produce tangible results. Examples of the actuators / motors 103 include fuel injections, windshield wipers, brake light circuits, transmissions, airbags, sensor-mounted motors (e.g., a motor configured to pivot a local sensor 102), motors, powertrain motors, steering, blind spot warning lights, etc.

[0013] The local sensors 102 transmit digital readings or measurements to the processors 108. Examples of local sensors 102 include temperature sensors, rotation sensors, seatbelt sensors, speed sensors, cameras, LiDAR sensors, radar sensors, infrared sensors, ultrasonic sensors, clocks, humidity sensors, rain sensors, light sensors, etc. It will be understood that any of the various electronic components of the Fig. It may include one separate or dedicated processor and memory. Further details regarding the structure and operations of the Computing System 100 are described, for example, in Miller, Prasad, Lavoie, and Hu.

[0014] Fig. Figure 2 shows and illustrates the carrier vehicle 200, which contains the computing system 100. Some of the local sensors 102 are mounted on the outside of the carrier vehicle 200 (others are located inside the vehicle 200). Local sensor 102a is configured to detect objects in front of the vehicle 200. Local sensor 102b is configured to detect objects behind the vehicle 200, as specified by the rear detection area 109b. The left sensor 102c and the right sensor 102d are configured to perform the same functions for the left and right sides of the vehicle 200, respectively.

[0015] As previously explained, the local sensors 102a to 102d can be ultrasonic sensors, LiDAR sensors, radar sensors, infrared sensors, cameras, microphones, and any combination thereof, etc. The carrier vehicle 200 incorporates a variety of other local sensors 102 located inside or outside the vehicle. The local sensors 102 can include one or all of the sensors disclosed in Miller, Prasad, Lavoie, D'Amato, and Hu. The general arrangement of the components, which are described in Fig. 1 and Fig. The state of the art shown in 2 is shown.

[0016] It is important to understand that the carrier vehicle 200, and more precisely the processors 108 of the carrier vehicle 200, is / are configured to perform the procedures and operations described below. In some cases, the carrier vehicle 200 is configured to perform these functions via computer programs stored in the volatile memory 107 and / or the non-volatile memory 106 of the computing system 100.

[0017] A processor is "configured" to perform a disclosed process step, block, or operation if at least one or more processors are in effective communication with a memory that stores a software program containing code or instructions embodying the disclosed process step or block. A further description of how the processors, memory, and software interact appears in Prasad. According to some embodiments, a mobile phone or external server, in effective communication with the carrier vehicle 200, performs some or all of the processes and operations described below.

[0018] According to various embodiments, the carrier vehicle 200 incorporates some or all of the functions of Prasad's vehicle 100a. According to various embodiments, the computing system 100 incorporates some or all of the functions of the VCCS 102. Fig. 2 of Prasad. According to various embodiments, the carrier vehicle 200 communicates with some or all of the in Fig. 1. Devices shown by Prasad, including the nomadic or mobile device 110, the transmission mast 116, the telecommunications network 118, the Internet 120, and the data processing center 122 (i.e., one or more servers). Each of the units described in this application (e.g., connected infrastructure, other vehicles, mobile phones, servers) may perform some or all of the functions described with reference to Fig. 1 and Fig. 2. Can be described, shared.

[0019] The term “equipped vehicle”, when used in the claims, is hereby defined as meaning: “a vehicle comprising: an engine, a plurality of wheels, a power source and a steering system; wherein the engine transmits torque to at least one of the plurality of wheels, thereby driving the at least one of the plurality of wheels; wherein the power source supplies energy to the engine; and wherein the steering system is configured to steer at least one of the plurality of wheels.” The carrier vehicle 200 can be an equipped vehicle.

[0020] The term “electrically equipped vehicle”, when used in the claims, is hereby defined as meaning “a vehicle comprising: a battery, a plurality of wheels, a motor, a steering system; wherein the motor transmits torque to at least one of the plurality of wheels, thereby driving the at least one of the plurality of wheels; wherein the battery is rechargeable and configured to supply the motor with electrical energy, thereby driving the motor; and wherein the steering system is configured to steer at least one of the plurality of wheels.” The carrier vehicle 200 can be an electrically equipped vehicle.

[0021] Fig. Figure 3 is a block diagram for generating driving decisions based on (a) captured external units and (b) captured environmental context (also referred to as context). Driving decisions include any instruction that involves a physical, tangible change in the carrier vehicle 200. Driving decisions include instructions to accelerate, decelerate (e.g., brake), reroute, change path (e.g., adjust course), and issue a warning (e.g., flash lights, generate sound). Driving decisions can also include any instruction that causes a physical, tangible change in an external vehicle. External units are physical, tangible external objects and include external vehicles, pedestrians, and obstacles (e.g., buildings, walls, potholes).Environmental context is typically non-physical and intangible (although it can be understood by referring to physical and tangible objects such as road signs, painted road markings, or non-drivable areas like grass). Environmental context thus represents rules chosen by humans in connection with driving. Examples of these rules include speed limits, restricted areas, assigned traffic flow directions, and designated stopping points (as indicated, among other things, by stop signs and red lights).

[0022] Context and units can be derived from the local sensors 102 and / or the telematics 104 (e.g., pre-generated street maps received from servers, external vehicles, and external infrastructure). While cameras may be configured to capture external units, the processing software required to convert images into coordinates of external units is inefficient and sometimes inaccurate. However, cameras are efficient at capturing contrast and color. The processing software associated with infrared sensors, radar sensors, LiDAR sensors, and / or ultrasonic sensors is efficient at converting sensor measurements into coordinates of external units. This processing software is inefficient at capturing color and contrast and sometimes unable to do so.In other words, cameras are better at capturing two-dimensional but not three-dimensional information, while infrared sensors, radar sensors, LiDAR sensors and / or ultrasonic sensors are better at capturing three-dimensional but not two-dimensional information.

[0023] While the environment is being measured, an image from a local camera sensor 102 may be insufficient due to limited ambient lighting or weather conditions. To enhance the robustness of the context measurement system, vector- or raster-based graphic data from a navigation system (consisting of stored map data, GPS, compass) can be forwarded to the image processing algorithms on board the carrier vehicle 200. These images can be used to refine the regions in which the image processing subsystem performs context analysis or to increase the probability of an environment classification used for braking / acceleration / path planning decisions.

[0024] In practice, context markers are often two-dimensional or effectively two-dimensional. For example, letters printed on a road sign are effectively two-dimensional. A computer is best able to distinguish printed letters from environmental noise by comparing the contrast and / or color of the printed letters with the surrounding environment. Similarly, painted lane markings are effectively two-dimensional. A computer is best able to distinguish painted lane markings from environmental noise by comparing the contrast or color of the painted lane markings with the surrounding environment. The carrier vehicle 200 can thus be configured to clarify environmental context using local camera sensors 102 and external (static or dynamic) units with local sensors that are not cameras, such as local sensors 102 with radar, LiDAR, or ultrasound.

[0025] With reference to Fig. 3. The carrier vehicle 200 extracts context 301 from sensors 102 and / or data received via the telematics 104 (e.g., a database including road speed limits and a road map). The carrier vehicle 200 extracts units 302 from local sensors 102 and / or data received via the telematics 104. Extracted units include properties such as position, two- or three-dimensional size and shape, speed, acceleration, course, and identity (e.g., animal, pedestrian, vehicle).

[0026] Once the external units have been identified, they are analyzed in light of the context to generate collision hazard assessments 303. Collision hazard assessments include, for example, time to collision and / or distance to collision, etc. Collision hazard assessments consider projected future properties of external units and / or the carrier vehicle 200. For example, a time-to-collision analysis might assume that an external unit traveling at a certain speed will maintain that speed, and the carrier vehicle 200 will maintain its current speed. The carrier vehicle 200 then generates driving decisions 304 based on the collision hazard assessment 303.

[0027] Methods for extracting units and properties from local sensors are known in the prior art. Methods for determining collision hazards are also known in the prior art. Some methods for determining collision hazards are disclosed in US patent application no. 15 / 183,355 by Bidner, which is hereby incorporated in its entirety by reference.

[0028] Fig. 4 is a more specific embodiment of the methods and operations which refer to Fig. 3. External units 401a to 401n transmit data via the telematics 104 to the carrier vehicle 200. Some external units (e.g., external unit 401b) can act as a proxy for unconnected external units (e.g., external unit 401a). The telematics 104 forwards the received data to the processors 108 and / or memory 106, 107. The local sensors 102 forward the acquired data (e.g., measurements) to the processors 108 and / or memory 106, 107.

[0029] The processors 108 create a virtual map 402 around the carrier vehicle 200 based on the forwarded data. The virtual map 402 need not be graphically represented or representable. For example, the virtual map 402 can be embodied as objects or their attributes stored in memory 106, 107. Suitable programs or software for constructing virtual maps based on acquired (i.e., received) data (such as road maps) are known in the prior art. The virtual map can be two- or three-dimensional. The virtual map includes clarified, recognized, or received units 403 arranged in clarified, recognized, or received context 404. Context includes some or all of the following: (A) Area location and identity, which includes (i) a non-drivable or closed area and (ii) a drivable area. A drivable area may be subdivided or divided by a speed limit. The carrier vehicle 200 determines drivable areas using received map information from external sources. The local context sensors 102 supplement this information using contrast and / or color from images. Image processing software clarifies a drivable area based on contrast and / or color and accordingly separates a drivable area from a non-drivable area. For example, an area that appears green and outside the lanes is marked as a non-drivable area. (B) Lanes of a drivable surface, which includes (i) the location of the lanes and (ii) the identity of the lanes. The location of the lanes includes some or all of the following: lane length, lane width, lane coordinates, lane curvature, and number of lanes. The identity of the lanes corresponds to the rules of the lanes. The rules include the direction of traffic flow and the legality of a lane change. The carrier vehicle 200 determines lanes and their properties using map information received from external sources. The local sensors 102 supplement this information using contrast and / or color from images. Image processing software clarifies lane lines based on contrast and / or color. The processors 108 determine any one of the above lane properties based on the clarified lane lines. (C) Parking spaces on the drivable surface, which includes (i) the location of the parking spaces and (ii) the identity of the parking spaces. The location of the parking space includes the width and depth of the parking spaces. The identity of the parking spaces includes rules associated with the parking area (e.g., only parallel parking allowed, only parking for disabled persons allowed). The carrier vehicle 200 determines parking spaces and their properties using map information received from external sources. The local sensors 102 supplement this information using contrast and / or color of images. Image processing software clarifies the boundaries of parking spaces (e.g., painted parking lines) based on the contrast and / or color of the images. The processors 108 determine any of the above lane properties based on the clarified boundaries.

[0030] As explained above, the Carrier Vehicle 200 applies the clarified context to clarified units. Applying clarified context to clarified units involves predicting or estimating future properties (e.g., position, velocity, acceleration, heading) of the clarified units based on the clarified context. Thus, the predicted or estimated future properties of the clarified units depend at least on (a) the current properties of the clarified units and (b) the context. Examples are provided below.

[0031] As described above, the carrier vehicle 200 performs a collision hazard assessment 405 of the clarified units based on their predicted or estimated future characteristics. As described above, the carrier vehicle 200 generates driving decisions 406 based on the collision hazard assessments 405.

[0032] Fig. Figure 5 illustrates an example virtual map of the carrier vehicle 200. A roundabout 501 intersects roads 502, 503, and 506. Road 502 is a one-way road in the direction from the carrier vehicle 200 to the roundabout center 501a. Road 506 intersects roads 505 and 504. The roundabout includes lanes 501c and 501d, separated by a lane line 501b and a non-drivable center 501a. Lanes 501c and 501d carry parallel traffic flow, as indicated by the dashed lane line 501b. Road 506 includes lanes 506b and 506c. Lane 506b carries traffic in the opposite direction to lane 506c, as indicated by the double lane line 506a. The carrier vehicle 200 is on road 502 at a speed of 200a (where speed includes velocity and course).A second vehicle, 201, is located in lane 501c of roundabout 501 at a speed of 200b. A third vehicle, 202, is located on road 504 at a speed of 202a. A fourth vehicle, 203, is located in lane 506b of road 506 at a speed of 203a. A fifth vehicle, 204, is located on road 505 at a speed of 204a.

[0033] As in Fig. As shown in Figure 5, each vehicle is represented by a box. Each box contains a triangle (unlabeled). Each triangle points towards the front bumper of the corresponding vehicle. In other figures, the triangles may be of different sizes. Such a change in size is not intended to convey any underlying meaning unless otherwise stated and is simply a drawing technique employed to improve clarity.

[0034] The carrier vessel speed 200a is zero, so vessel 200 is stationary. The second speed 201a includes a course 201a pointing towards the carrier vessel 200. Therefore, if the carrier vessel 200 remains stationary and if the second vessel 201 were to continue along the second course 201a (as noted above, speed includes course), the carrier vessel 200 and the second vessel 201 would collide.

[0035] However, lane 501c is curved. If the second vehicle 201 follows lane 501c, the second course 201a turns to remain parallel to lane 501c. Based on the context (i.e., the curvature of lane 501c), the carrier vehicle 200 extrapolates that the second course 201a follows lane 501c. A collision risk assessment between carrier vehicle 200 and the second vehicle 201 yields a zero or low probability of collision. Consequently, the carrier vehicle 200 does not generate a driving decision (e.g., an evasive maneuver) based on an expected collision between carrier vehicle 200 and the second vehicle 201. According to some embodiments, driving decisions are based on a magnitude of the collision probability.

[0036] Fig. 6 and Fig. 7 refer to first embodiments of a collision hazard assessment. Fig. Sections 8 to 14 refer to second embodiments of a collision hazard assessment. Features of the first and second embodiments can be combined.

[0037] With reference to Fig. 6. The carrier vehicle 200 reacts to a trigger by calculating a reference segment 601a extending from the second vehicle 201 to the carrier vehicle 200. The trigger is based on the position and velocity of the carrier vehicle 200 and the position and velocity of the second vehicle 201.

[0038] In Fig. In section 6, the reference segment 601a is the shortest segment connecting the second vehicle 201 to the carrier vehicle 200. According to other embodiments, the reference segment 601a extends from a midpoint of the carrier vehicle 200 to a midpoint of a front surface of the vehicle 201. The carrier vehicle 200 calculates a series of curved segments 601b to 601g that intersect both ends of the reference segment 601a. ​​The curved segments 601b to 601g may be spaced at predetermined intervals. The curved segments 601b to 601g follow predetermined geometric functions. Some or all of them may be parabolic. A total number of curved segments 601b to 601g, calculated from the first vehicle 200, is based on (a) the predetermined distances and (b) the speed and / or acceleration of the carrier vehicle 200 and / or the second vehicle 201.A number of the curved sections on each side of the reference section 601a are based on the speed and / or acceleration of the carrier vehicle 200 and / or the second vehicle 201.

[0039] The outer curved segments 601f and 601g correspond to extreme paths of the second vehicle 201. Based on the second velocity 201a, the outer curved segments 601f and 601g represent, for example, the most extreme curved collision paths between the second vehicle 201 and the carrier vehicle 200, which would not cause the second vehicle 201 to become uncontrollable (e.g., skid or roll over). The carrier vehicle 200 is pre-equipped with one or more functions that determine the curvature of the outer curved segments 601f and 601g based on properties of the second vehicle 201 (e.g., velocity).

[0040] According to one embodiment, after determining the reference segment 601a, the carrier vehicle 200 determines the outer curved segments 601f and 601g and then establishes a first predetermined number of preliminary curved segments 601c and 601e between the outer curved segment 601g and the reference segment 601a and a second predetermined number of preliminary curved segments between the outer curved segment 601f and the reference segment 601a. ​​The first and second predetermined numbers can be (a) pre-defined, (b) equal, or (c) based on an angle of the reference segment 601a with respect to the second velocity 201a.

[0041] The carrier vehicle 200 evaluates each segment 601 in light of the context. For each segment 601, the carrier vehicle 200 can determine a number of rules (derived from the context, as described above) that will be broken by the second vehicle 201. The carrier vehicle 200 can also determine the extent of each broken rule.

[0042] If the second vehicle 201 were to follow segment 601f, the second vehicle 201 would (a) illegally exit roundabout 501 at angle 602, (b) illegally cross the non-drivable area 507 for a distance defined between points 603 and 604, and (c) illegally enter road 502 at angle 605. Angle 602, the distance between points 603 and 604, and angle 605 correspond to the extent of the broken rules.

[0043] Instead of identifying each prohibited act, the carrier vehicle 200 can calculate segments of each sub-segment 601 that correspond to one or more violated rules. For example, the reference sub-segment 601a can be subdivided into a permitted first segment extending from the second vehicle 201 to point 608, and a prohibited second segment extending from point 608 to the carrier vehicle 200. While traversing the first segment of sub-segment 601a, the second speed 201a would sufficiently conform to the curvature of lane 501c (i.e., conform to predetermined limits) to be considered permitted. While traversing the second segment of sub-segment 601a, the second speed 201a would sufficiently deviate from the curvature of lane 501c and ultimately road 502 to be considered prohibited.Thus, the extent of the unauthorized activity on section 601a would be related, at least in part, to the distance of the second section. In contrast, the entire section 601g would qualify as unauthorized, since the second speed 201a would deviate sufficiently from the curvature of lane 501c at any point along section 601g and would ultimately be contrary to the direction of travel on road 502 (as explained above, road 502 is a one-way road towards the center of 501a).

[0044] The carrier vehicle 200 performs collision risk assessments for each segment 601 in light of one or more of the following: whether the segment 601 involves a prohibited activity (i.e., an activity that violates the context) and the degree or extent of the prohibited activity. Segments 601 with a greater degree of prohibited activity are disregarded or considered less likely. The carrier vehicle 200 sums the collision risk assessments for each segment 601. If the sum exceeds a predetermined probability threshold, the carrier vehicle 200 generates a driving decision (e.g., controlling the steering, braking, and / or acceleration) corresponding to an evasive maneuver (i.e.,a maneuver that is calculated to (a) reduce the probability of a collision and / or (b) reduce a likely differential speed between carrier vehicle 200 and second vehicle 201 in the event of a collision).

[0045] Referring to Fig. Vehicle 200 travels along a two-lane one-way street 707 at a speed of 200a. The second vehicle, 201, enters a parking space 703 in parking lot 702 at a speed of 201a. The third, fourth, and fifth vehicles, 202, 203, and 204, are already parked. Parking space 703 is defined by two side lines (unlabeled) and an end line 704. A concrete barrier 705 separates street 707 from parking lot 702. Street 707 includes a painted line 706 adjacent to and extending parallel to street 707.

[0046] The carrier vehicle 200 clarifies vehicles 201 to 204 using local unit sensors and assigns them speeds. The carrier vehicle 200 determines the second speed 201a (as explained above, speeds include velocity and course). The carrier vehicle 200 determines the carrier vehicle speed 200a. The carrier vehicle 200 clarifies the context, at least partially, using local context sensors 102a.

[0047] In one instance, the carrier vehicle 200 identifies 200 painted lines representing seven parking spaces using image processing software. The carrier vehicle 200 compares the identified painted lines with pre-loaded reference parking space geometries. The carrier vehicle 200 identifies that the width between the painted lines (corresponding to the width of the parking spaces) lies within a predetermined range of widths in the pre-loaded reference parking space geometries. The carrier vehicle 200 identifies that each parking space is defined by three painted lines and is therefore an open rectangle, which matches at least some of the pre-loaded reference parking space geometries. The carrier vehicle 200 identifies that the parking spaces are grouped into a multitude of adjacent parking spaces. The carrier vehicle applies a parking space context to parking space 703 if some or all of these identifications are present.The carrier vehicle 200 confirms the parking space context with map information received from an external server, which identifies the area associated with parking space 702 as a parking space.

[0048] In another case, the carrier vehicle 200 receives (or has previously received) information from an external server that identifies the coordinates of parking space 702. Based on this previously received information, the carrier vehicle 200 scans the coordinates of parking space 702 with the local sensors 102 and confirms that the received information is consistent with features of images captured by the local sensors 102.

[0049] The concrete wall 705 is three-dimensional and can be clarified using the local sensors 102. The carrier vehicle 200 marks the concrete wall 705 as stable infrastructure and identifies that the height of the concrete wall 705 exceeds the vertical coordinates of the parking space 702.

[0050] The carrier vehicle 200 performs a collision risk assessment between the second vehicle 201 and the carrier vehicle 200. If the carrier vehicle 200 continues at carrier vehicle speed 200a and the second vehicle 201 continues at the second speed 201a, the carrier vehicle 200 and the second vehicle 201 would collide at point 701.

[0051] However, the carrier vehicle 200 projected future positions and speeds of the second vehicle 201 based on the context. More precisely, the carrier vehicle 200 projected that the second vehicle's speed 201a would decrease in light of the end line 704 of parking space 703. The carrier vehicle 200 projected that a future position of the second vehicle 201 would respect (i.e., not cross) the end line 704. The carrier vehicle 200 performed a similar analysis with respect to the concrete barrier 705 and the painted line 706. It is therefore important to understand that when performing collision hazard assessments, the carrier vehicle 200 projected future speeds of units based on the clarified context.

[0052] According to some embodiments, the trigger for extending the reference section 601a between the carrier vehicle 200 and the second vehicle 201 is based on the absence of any units (including infrastructure, such as the concrete barrier 705) that obstruct the collision paths of the second vehicle 201 with the carrier vehicle 200. Units obstruct collision paths if they are solid, have at least a predetermined thickness, intersect a ground plane with a predetermined angular range, and terminate at least a predetermined height above the ground adjacent to the infrastructure on the side closest to the second vehicle 201. Here, the concrete barrier 705 exceeds the predetermined thickness, intersects the ground at 90 degrees (and is thus within the predetermined angular ranges), and extends a predetermined height above the ground between the end line 704 and the concrete barrier 705.Thus, the carrier vehicle 200 does not extend the reference section 601a between carrier vehicle 200 and second vehicle 201.

[0053] Fig. Sections 8 to 14 refer to second embodiments of a collision hazard assessment. These embodiments may include any features that were explained with reference to the first embodiments of the collision hazard assessment. The processes of Fig. 8 to 14 can be applied to any collision risk assessment, not just a cross-traffic warning, and to any traffic situation (e.g., not just when the carrier vehicle is reversing into a road).

[0054] With reference to Fig. From 8 to 14, the carrier vehicle 200 can be configured to issue cross-traffic (CT) warnings, which represent a form of collision risk assessment. CT warnings alert the carrier vehicle driver if the carrier vehicle 200 might reverse into cross traffic. With reference to Fig. 8. The CT warnings are based on local sensors 102. Rear local sensors 102b may include local CT sensors 102e and 102f, each configured to emit CT signal patterns 801 and 802, respectively. The local CT sensors 102e and 102f may be radar, LiDAR, ultrasound, or another type of sensor, as previously explained in relation to the local sensors 102.

[0055] The carrier vehicle 200 (and more precisely the processors 108, as previously explained) is / are configured to extrapolate a path of the external units detected by the local CT sensors. The general concept of CT alerts is known in the prior art.

[0056] With reference to Fig. 8. It is extrapolated that the second vehicle 201 will intersect (i.e., collide with) the carrier vehicle 200, based on the position, speed, and / or acceleration of the second vehicle 201 as detected by the local CT sensors 102e, 102f, and based on the position, speed, and / or acceleration of the carrier vehicle 200 as detected by other local sensors 102. As a result, the carrier vehicle 200 issues a CT warning. The CT warning may include a warning displayed on a user interface 105 (including a flashing light), automatic application of the brakes, etc. Further, with reference to Fig. 8 determines the carrier vehicle 200, even though a third vehicle 202 is detected and has stopped. As a result, the carrier vehicle can ignore the third vehicle 202.

[0057] With reference to Fig. 9. Carrier vehicle 200 is configured to apply virtual map 402 to improve the accuracy of CT alerts. In Fig. 9 the carrier vehicle 200 is on an entrance 802 and is reversing into lane 501c. Fig. 9 is otherwise similar to Fig. 5. Due to at least some existing CT warning systems, the carrier vehicle 200 would incorrectly extrapolate the path of the second vehicle 201 as path 901. Path 901 represents the path of the second vehicle 201 if the current second speed 201a were to continue in the future. However, it is more likely that the second vehicle 201 will follow the curve of lane 501c, and thus path 902 is more likely than path 901. Therefore, with at least some existing CT warning systems, the carrier vehicle 200 would incorrectly extrapolate that stopping at its current position would result in a collision and reversing into lane 501c would avoid a collision.However, the correct projection (or at least the projection that is most likely correct) is that if vehicle 200 stopped immediately in entrance 802, it would avoid a collision, whereas reversing into lane 501c (and thus cutting off path 902) would result in a collision.

[0058] The carrier vehicle 200 is configured to achieve the correct result by (a) recognizing the current properties of the second vehicle 201 and (b) modifying these current properties in light of the context. More precisely, and with reference to Fig. 10, the carrier vehicle 200 goes through some or all of the following operations: first, the current properties (including current lane) of the second vehicle 201 are recognized; second, based on the virtual map corresponding to the current lane of the second vehicle 201, an ideal path 1001 and / or a feasible path 1301 of the second vehicle 201 is predicted.

[0059] To find the ideal path 1001, the coordinates of a series of points 1002 can be found. The identification of the coordinates is carried out with reference to Fig. 11 explained. The paths explained above and below can represent the location of a fixed point on the second vehicle 201 (e.g., the position of the center of the front bumper of the second vehicle 201).

[0060] As in Fig. As shown in Figure 11, the current lane of the second vehicle 201 can be divided into segments 1111. Each segment 1111 can be defined as existing between the outer and inner lane boundaries 1112 and 1113, which define lane 501c (as identified in virtual map 402). Each segment 1111 can be defined as having the same (e.g., approximately the same) area. Each segment 1111 can be defined as having any area that is smaller than a predetermined area of ​​the segment.

[0061] Since entrance 502 intersects lane 501c, an additional boundary 1114 can be applied to separate entrance 502 from lane 501c. The additional boundary 1114 can be a straight line defined between opposite ends 1115 and 1116 of the outer boundary 1112. The additional boundary 1114 may already exist in the virtual map. The additional boundary 1114 can be curved with a curvature interpolated between the shared sections of the outer boundary 1112.

[0062] Each segment can be defined between the outer and inner boundaries 1112, 1113 and transverse boundaries 1117. Each transverse boundary 1117 can be defined such that it intersects both boundaries 1112 and 1113 at angles 1119, 1120 within a predetermined range (e.g., ± 10%) of 90 degrees. As in Fig. As shown in Figure 11, the transverse boundary 1117b of a first segment 1111a can serve as the transverse boundary 1117b of an adjacent segment 1111b. The process described above can be repeated, with continuous remeasuring of the segments 1111 (and thus repositioning of the transverse boundaries 1117), until the conditions described above are met. Then, a midpoint 1122 of each transverse boundary 1117 is found. The ideal path 1001 is then interpolated based on the midpoints 1122 (e.g., the ideal path 1001 could be the best-fitting line that intersects each midpoint 1122).

[0063] To compensate for real-world conditions (e.g., the second vehicle 201 deviates from the ideal path 1001), the ideal path 1001 can be widened to form a practical path 1301. To find the practical path 1301, a set of practical outer and inner points 1201, 1202 can be determined. The practical outer points 1201 can be defined such that they lie on the transverse boundaries 1117 at a predetermined distance outside the outer boundary 1113 or outside the ideal path 1001. The practical inner points 1202 can be defined such that they lie on the transverse boundaries 1117 at a predetermined distance inside the boundary 1113 or inside the ideal path 1001.

[0064] Returning to Fig. 9. The path 902 of the second vehicle 201 can thus be defined as a practical path 1301 or as an ideal path 1001 (depending on the embodiment used). The carrier vehicle 200 can apply any of the operations described above or any other suitable operation to predict a carrier path (and associated times or spans). The carrier vehicle 200 determines whether the carrier path (not shown) intersects the second vehicle path 902. If an intersection occurs, the carrier vehicle 200 determines whether the intersection occurs simultaneously (e.g., at a single time or within a certain overlap span of that single time). One must therefore, with reference to Fig. 9 understand that the carrier vehicle 200 may not extrapolate a collision if the path of the carrier vehicle 200 does not intersect the path of the second vehicle 201, if the second vehicle 201 were to follow lane 501d instead of lane 501c, and if it were extrapolated that the carrier vehicle 200 only occupies entrance 802 and lane 501c.

[0065] To find a simultaneous overlap, the carrier vehicle 200 can cycle through a series of time overlap intervals. At each time interval, it can be determined whether the predicted occupied area of ​​the carrier vehicle 200 intersects the predicted occupied area of ​​the second vehicle 201 (or occurs within a predetermined distance thereof). The predicted occupied areas are at least the dimensions of the respective vehicles and can be larger, since, as explained below, it can be predicted that each vehicle can occupy a range of locations at any given time.

[0066] To account for the timing of the second vehicle 201, each lateral boundary 1117 can be assigned to the front center of a front bumper of the second vehicle 201 and can be linked to a time point (or a range of time points) based on current characteristics of the second vehicle 201 (e.g., speed, acceleration). If ranges are applied, then the ranges become wider as the distance from the current position of the second vehicle 201 increases.

[0067] As an example, the first transverse boundary 1117a can be linked to a future time of one second, and the second transverse boundary 1117b can be linked to a future time of three seconds. As another example, the first transverse boundary 1117a can be linked to a future time of one to four seconds (a span of three seconds), the second transverse boundary 1117b can be linked to a future time of three to seven seconds (a span of four seconds), and a third transverse boundary (unlabeled) can be linked to a future time of five to eleven seconds (a span of six seconds). The times associated with the inner regions of the segments 1111 can be interpolated between successive transverse boundaries 1117. In parallel with these operations, the timing of the carrier vehicle path is also determined using the same or another suitable technique.

[0068] Each overlap time interval is assigned to any position of the second vehicle 201 and the carrier vehicle 200 according to the time interval. For example, if the overlap time interval is 0.1 seconds, then every position of the second vehicle that occurs at 0.1 seconds is identified as corresponding. Imagine that the first transverse boundary has a time span of 0.01 seconds to 0.2 seconds and that the second transverse boundary has a time span of 0.1 seconds to 0.4 seconds. According to this example, all first segments 1111a would correspond to the overlap time interval of 0.1 seconds. Thus, an occupied area of ​​the second vehicle 201 at an overlap time interval of 0.1 seconds would include all first segments 1111a.

[0069] To account for the body of the second vehicle 201, which has a two-dimensional area, any occupied area can be extended. For example, if the path is assigned to the center of the front bumper of the second vehicle 201, the occupied area can be extended towards the current position of the second vehicle 201 to account for the body of the second vehicle 201. As an example and with reference to Fig. 14. The corresponding span of positions 1401 at a third time interval of 0.3 seconds (0.1*3, where 3 represents the time interval) can be all of the first segments 1111a and a section of the second segment 1111b. A body area 1402 is added to the corresponding span of positions 1401 to generate a total occupied area 1403 equal to the area of ​​1401 plus the area of ​​1402. For the reasons explained above, the total occupied area increases as the time interval progresses. For example, at a time interval of 0.1 seconds, the total occupied area can be 100 m. 2 The next time interval of 0.2 seconds can be the entire occupied area of ​​150 m. 2 The next time interval of 0.3 seconds can cover the entire occupied area of ​​250 m. 2 be.

[0070] The same or other suitable operations are performed for carrier vehicle 200. If the occupied area of ​​carrier vehicle 200 and the second vehicle 201 overlap (or occur within a predetermined distance of each other) at a given time overlap interval, then a simultaneous overlap is determined. Progression through the time intervals can only be performed until a simultaneous overlap is determined (i.e., it ends as soon as a simultaneous overlap is determined). If a simultaneous overlap exists over a given time interval, then the calculation ends and a CT warning is immediately issued. If no simultaneous overlap exists, then the calculation can proceed to the next time interval.

[0071] As explained above, the references to Fig. Operations 8 to 14 described above can be applied to non-CT warning systems (e.g., other forms of collision hazard assessment). These operations, with reference to the Fig. The operations described in sections 8 to 14 can be applied to any of the embodiments described above, including all those described with reference to Fig. 1 to 7 described embodiments.

Claims

[1] Carrier vehicle, comprising: a motor, brakes, sensors and a processor, configured to: to identify a lane of the target vehicle; to determine a radius of curvature of the road boundaries of the detected lane; to predict a target path of the target vehicle based on lane boundaries of a virtual map and the target path based on the specified radius of curvature; to compare the target path with a predicted carrier path of the carrier vehicle; and to apply the brakes based on the comparison. [2] Carrier vehicle according to claim 1, wherein the processor is configured to create the virtual map based on (a) a received road map and (b) measurements received from the sensors. [3] Carrier vehicle according to claim 2, wherein the sensors comprise a camera and the processor is configured to display the lane boundaries on the virtual map based on images captured by the camera [4] Carrier vehicle according to claim 1, wherein the processor is configured to (a) determine a first radius of curvature of a first lane boundary of the detected lane, (b) determine a second radius of curvature of a second lane boundary of the detected lane and an intermediate stage; calculate an intermediate radius of curvature based on (a) and (b); predict the target path based on the calculated intermediate radius of curvature. [5] Carrier vehicle according to claim 4, wherein the processor is configured to predict, based on the target path, a range of positions of the target vehicle at a first future time and a range of positions of the target vehicle at a second future time. [6] Carrier vehicle according to claim 5, wherein the first future time is a predetermined time interval multiplied by a first number, wherein the second future time is the predetermined time interval multiplied by a value, and wherein the value is the first number plus one. [7] Carrier vehicle according to claim 6, wherein the processor is configured such that a total range of the predicted span of positions of the target vehicle at the second future time exceeds a total range of the predicted span of positions of the target vehicle at the first future time. [8] Carrier vehicle according to claim 7, wherein the processor is configured such that a total range of the predicted range of positions of the target vehicle at the first future time exceeds a total range of the target vehicle. [9] Carrier vehicle according to claim 1, wherein the processor is configured to predict the target path and the carrier path as shapes, each having a two-dimensional surface area. [10] Carrier vehicle according to claim 9, wherein the processor is configured to determine whether the shapes intersect; and in response to determining whether the shapes intersect at an intersection point, to calculate a first time interval in which the target vehicle reaches the intersection point and a second time interval in which the carrier vehicle reaches the intersection point. [11] Carrier vehicle according to claim 10, wherein the processor is configured to determine whether any section of the first time span overlaps any section of the second time span. [12] Carrier vehicle according to claim 11, wherein the processor is configured to apply the brakes based on a positive overlap determination. [13] Carrier vehicle comprising: an engine, a steering system, sensors and a processor, configured to: to identify a lane of the target vehicle; to determine a radius of curvature of the road boundaries of the detected lane; to predict a target path of the target vehicle based on lane boundaries of a virtual map and the target path based on the specified radius of curvature; to compare the target path with a predicted carrier path of the carrier vehicle; and to operate the steering system based on the comparison. [14] Carrier vehicle according to claim 13, wherein the processor is configured to: to predict the target path and the carrier path as shapes, each of which has a two-dimensional area. to determine whether the shapes intersect and, based on this determination, to calculate an initial time period in which the target vehicle reaches the point of intersection, and to calculate a second time period in which the carrier vehicle reaches the intersection point. to determine whether any section of the first time period overlaps any section of the second time period. to operate the steering system based on a positive overlap determination. to create the virtual map based on (a) a received street map and (b) measurements received from the sensors.

Images