Method, device, vehicle and storage medium for autonomous driving vehicle to pass through intersection

Abstract

This application relates to the field of autonomous driving technology, and in particular to a method for autonomous vehicles to navigate intersections. The method includes the following steps: identifying whether a target vehicle exists within a preset range of the target intersection; when a target vehicle is identified, predicting the actual collision level based on the motion states of the vehicle and the target vehicle, and matching a first expected deceleration of the vehicle with a second expected deceleration of the target vehicle based on the actual collision level; when the first expected deceleration is greater than a first deceleration threshold and the second expected deceleration is greater than a second deceleration threshold, controlling the vehicle to accelerate or maintain its current speed to pass through the target intersection using a preset acceleration strategy or the current driving strategy; otherwise, controlling the vehicle to decelerate to pass through the target intersection using a preset deceleration strategy. This solves problems such as low vehicle driving safety, low traffic efficiency, lack of flexibility, and instability.

Description

Technical Field

[0001] The present application relates to the field of autonomous driving technology, and in particular to a method, device, vehicle and storage medium for an autonomous driving vehicle to pass through an intersection. Background Art

[0002] As the application areas of autonomous vehicles continue to grow, the driving scenarios that need to be handled are also gradually increasing. Take intersections as an example. When crossing an intersection or turning left / right at an intersection, you will often encounter people or motor vehicles or non-motor vehicles with uncertain movement directions. If you cannot avoid them in time, it will cause traffic accidents. If you just avoid them, it will cause traffic congestion.

[0003] In related technologies, autonomous driving vehicles mainly determine whether they are in a meeting scene and whether there is a risk of collision when meeting other vehicles through scene recognition and target trajectory prediction methods. They only consider collision analysis of the interaction between the vehicle and the driving trajectory in a meeting scene, resulting in low vehicle driving safety, low traffic efficiency, lack of flexibility and instability. Summary of the invention

[0004] The present application provides a method, device, vehicle and storage medium for an autonomous driving vehicle to pass through an intersection, in order to solve the problems in the related art that autonomous driving vehicles usually pass through intersections using conservative or aggressive strategies, which cannot take into account the safety and efficiency of intersection passage and result in poor user experience.

[0005] In a first aspect, an embodiment of the present application provides a method for an autonomous driving vehicle to pass through an intersection, comprising the following steps: identifying whether a target vehicle exists within a preset range of the target intersection; when the target vehicle is identified, predicting an actual collision level based on the motion states of the host vehicle and the target vehicle, and matching a first expected deceleration of the host vehicle with a second expected deceleration of the target vehicle based on the actual collision level; when the first expected deceleration is greater than a first deceleration threshold, and the second expected deceleration is greater than a second deceleration threshold, controlling the host vehicle to accelerate or maintain the current speed to pass through the target intersection using a preset acceleration strategy or a current driving strategy, otherwise controlling the host vehicle to decelerate and pass through the target intersection using a preset deceleration strategy.

[0006] According to the above-mentioned technical means, the embodiments of the present application can determine the actual collision risk between the vehicle and the target vehicle through the expected deceleration. The greater the expected deceleration, the higher the corresponding collision risk. The reasonable intersection traffic strategy is determined by comprehensively considering the collision risk between the vehicle and the target vehicle. When the collision risk is low, an active traffic strategy can be used to pass through the intersection to ensure the efficiency of intersection traffic. When the collision risk is high, a conservative traffic strategy can be used to pass through the intersection to ensure the safety of intersection traffic. This can take into account both the safety and efficiency of intersection traffic and improve user experience.

[0007] Furthermore, predicting the actual collision level based on the motion states of the host vehicle and the target vehicle includes: predicting whether there is an overlapping area in the preset safety ranges of the host vehicle and the target vehicle respectively based on the motion states of the host vehicle and the target vehicle; if there is an overlapping area, matching the actual collision level according to the area of ​​the overlapping area, wherein the area of ​​the overlapping area is positively correlated with the actual collision level, and a larger actual collision level indicates a higher collision risk; if there is no overlapping area, controlling the host vehicle to maintain the current speed through the target intersection according to the current driving strategy.

[0008] According to the above technical means, the embodiment of the present application can quickly and accurately determine the degree of collision risk based on the degree of overlap between the safety range of the own vehicle and the target vehicle, so as to facilitate the determination of subsequent intersection traffic strategies.

[0009] Furthermore, before identifying whether there is a target vehicle within a preset range of the target intersection, it includes: acquiring environmental information and / or positioning information around the vehicle; identifying the environmental information to obtain a first intersection identifier, and / or matching a second intersection identifier based on the positioning information and a preset map; and determining the target intersection based on the first intersection identifier and / or the second intersection identifier.

[0010] According to the above technical means, the embodiments of the present application can use a variety of methods to determine the intersection signs, so as to facilitate the determination of the target intersection and effectively improve the feasibility and applicability of the target intersection identification.

[0011] Furthermore, the target intersection is determined based on the first intersection sign and the second intersection sign, including: calculating a first confidence of the second intersection sign based on the positioning information, and calculating a second confidence of the first intersection sign based on the environmental information; when the first confidence is greater than a first confidence threshold, determining the target intersection based on the second intersection sign; when the first confidence is less than or equal to the first confidence threshold, and the second confidence is greater than the second confidence threshold, determining the target intersection based on the first intersection sign, otherwise, outputting no intersection information.

[0012] According to the above-mentioned technical means, in the embodiment of the present application, a variety of information can be integrated to determine the target intersection, avoid misidentification of the target intersection due to the use of a single method, and improve the accuracy of target intersection identification, thereby ensuring stable driving of the vehicle by improving the accuracy of the information and ensuring a stable source of information.

[0013] Further, the calculating the first confidence of the second intersection sign according to the positioning information includes: establishing a positioning covariance matrix according to the positioning information; and obtaining the first confidence by matching the variance value and the covariance value of the positioning covariance matrix.

[0014] According to the above technical means, in the embodiment of the present application, a positioning covariance matrix is ​​established based on the positioning information, and the first confidence level is obtained by matching the variance value and covariance value of the positioning covariance matrix, so that the credibility of the second intersection sign can be accurately determined, thereby improving the user experience.

[0015] Further, calculating the second confidence of the first intersection sign according to the environmental information includes: identifying the environmental information to obtain actual environmental brightness; and obtaining the second confidence according to matching of the actual environmental brightness.

[0016] According to the above technical means, in the embodiment of the present application, the credibility of the first intersection sign is quickly and accurately determined based on the ambient brightness, thereby improving the user experience.

[0017] Furthermore, before identifying whether there is a target vehicle within a preset range of the target intersection, it includes: when the target intersection is identified, controlling the speed of the vehicle to decrease to a target speed, and controlling the vehicle to perform a target reminder action.

[0018] According to the above-mentioned technical means, in the embodiment of the present application, when the target intersection and the target vehicle are identified, the vehicle is controlled to slow down or a target reminder action is performed. For example, the high and low headlights are switched to remind the target vehicle or pedestrian approaching the intersection, and the human-computer interaction in the vehicle reminds the driver that he is about to approach the intersection, ensuring that when encountering the risk of accidental collision, the driver can take over the vehicle in time, thereby improving the safety, stability and flexibility of the vehicle.

[0019] In a second aspect, an embodiment of the present application provides an intersection passage device for an autonomous driving vehicle, comprising: an identification module for identifying whether a target vehicle exists within a preset range of the target intersection; a matching module for predicting an actual collision level based on the motion states of the host vehicle and the target vehicle when the target vehicle is identified, and matching a first expected deceleration of the host vehicle with a second expected deceleration of the target vehicle according to the actual collision level; a control module for controlling the host vehicle to accelerate or maintain the current speed through the target intersection according to a preset acceleration strategy or a current driving strategy when the first expected deceleration is greater than a first deceleration threshold and the second expected deceleration is greater than a second deceleration threshold, otherwise controlling the host vehicle to decelerate through the target intersection according to a preset deceleration strategy.

[0020] Furthermore, the matching module is used to: predict whether there is an overlapping area between the preset safety ranges of the host vehicle and the target vehicle according to the motion states of the host vehicle and the target vehicle; if there is an overlapping area, match the actual collision level according to the area of ​​the overlapping area, wherein the area of ​​the overlapping area is positively correlated with the actual collision level, and the larger the actual collision level, the higher the collision risk; if there is no overlapping area, control the host vehicle to maintain the current speed through the target intersection according to the current driving strategy.

[0021] Furthermore, it also includes: a determination module, which is used to obtain environmental information and / or positioning information around the vehicle before identifying whether there is a target vehicle within a preset range of the target intersection; identifying the environmental information to obtain a first intersection identifier, and / or matching a second intersection identifier based on the positioning information and a preset map; and determining the target intersection based on the first intersection identifier and / or the second intersection identifier.

[0022] Furthermore, the determination module is further used to: calculate a first confidence level of the second intersection sign based on the positioning information, and calculate a second confidence level of the first intersection sign based on the environmental information; when the first confidence level is greater than a first confidence threshold, determine the target intersection based on the second intersection sign; when the first confidence level is less than or equal to the first confidence threshold, and the second confidence level is greater than the second confidence threshold, determine the target intersection based on the first intersection sign, otherwise, output no intersection information.

[0023] Furthermore, the determination module is further used to: establish a positioning covariance matrix according to the positioning information; and obtain the first confidence by matching the variance value and covariance value of the positioning covariance matrix.

[0024] Furthermore, the determination module is further used to: identify the environmental information to obtain the actual environmental brightness; and obtain the second confidence level according to the actual environmental brightness matching.

[0025] Furthermore, the control module is further used to: before identifying whether there is a target vehicle within a preset range of the target intersection, when the target intersection is identified, control the speed of the vehicle to be reduced to a target speed, and control the vehicle to perform a target reminder action.

[0026] The third aspect of the present application provides a vehicle, comprising: a memory, a processor, and a computer program stored on the memory and executable on the processor, wherein the processor executes the program to implement the intersection passage method of the autonomous driving vehicle as described in the above embodiment.

[0027] The fourth aspect of the present application provides a computer-readable storage medium having a computer program stored thereon, which is executed by a processor to implement the intersection passage method of an autonomous driving vehicle as described in the above embodiments.

[0028] Therefore, this application has at least the following beneficial effects:

[0029] (1) The embodiment of the present application can determine the actual collision risk between the host vehicle and the target vehicle through the expected deceleration. The greater the expected deceleration, the higher the corresponding collision risk. The reasonable intersection traffic strategy is determined by comprehensively considering the collision risk between the host vehicle and the target vehicle. When the collision risk is low, an active traffic strategy can be adopted to pass through the intersection to ensure the traffic efficiency of the intersection. When the collision risk is high, a conservative traffic strategy can be adopted to pass through the intersection to ensure the safety of the intersection. In this way, the safety and efficiency of the intersection can be taken into account, thereby improving the user experience.

[0030] (2) The embodiment of the present application can quickly and accurately determine the degree of collision risk based on the degree of overlap between the safety range of the host vehicle and the target vehicle, so as to facilitate the determination of subsequent intersection traffic strategies;

[0031] (3) The embodiments of the present application can use a variety of methods to determine the intersection sign, so as to facilitate the determination of the target intersection and effectively improve the feasibility and applicability of the target intersection identification;

[0032] (4) In the embodiment of the present application, a variety of information can be integrated to determine the target intersection, thereby avoiding misidentification of the target intersection due to a single method, and improving the accuracy of target intersection identification, thereby ensuring stable driving of the vehicle by improving the accuracy of information and ensuring a stable information source;

[0033] (5) In the embodiment of the present application, a positioning covariance matrix is ​​established based on the positioning information, and the first confidence level is obtained by matching the variance value and the covariance value of the positioning covariance matrix, so that the credibility of the second intersection sign can be accurately determined, thereby improving the user experience;

[0034] (6) In the embodiment of the present application, the credibility of the first intersection sign is determined quickly and accurately based on the ambient brightness, thereby improving the user experience;

[0035] (7) In the embodiments of the present application, when a target intersection and a target vehicle are identified, the vehicle is controlled to slow down or a target reminder action is performed. For example, the high and low headlights are switched to remind the target vehicle or pedestrian approaching the intersection, and the human-computer interaction in the vehicle reminds the driver that the intersection is approaching, so as to ensure that the driver can take over the vehicle in time when encountering the risk of accidental collision, thereby improving the safety, stability and flexibility of the vehicle.

[0036] Additional aspects and advantages of the present application will be given in part in the description below, and in part will become apparent from the description below, or will be learned through the practice of the present application. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] The above and / or additional aspects and advantages of the present application will become apparent and easily understood from the following description of the embodiments in conjunction with the accompanying drawings, in which:

[0038] Figure 1 A flowchart of a method for passing through an intersection of an autonomous driving vehicle according to an embodiment of the present application;

[0039] Figure 2 A schematic diagram of a preset range of a method for passing through an intersection of an autonomous driving vehicle according to an embodiment of the present application;

[0040] Figure 3 A scene recognition logic block diagram of an intersection passage method for an autonomous driving vehicle according to an embodiment of the present application;

[0041] Figure 4 A target collision analysis diagram of an intersection passage method for an autonomous driving vehicle according to an embodiment of the present application;

[0042] Figure 5 A system framework diagram of a driving scenario of a method for passing through an intersection of an autonomous driving vehicle according to an embodiment of the present application;

[0043] Figure 6 This is an example diagram of an intersection passage device for an autonomous driving vehicle according to an embodiment of the present application;

[0044] Figure 7 It is a schematic diagram of the structure of a vehicle according to an embodiment of the present application. DETAILED DESCRIPTION

[0045] Embodiments of the present application are described in detail below, and examples of the embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals throughout represent the same or similar elements or elements having the same or similar functions. The embodiments described below with reference to the accompanying drawings are exemplary and are intended to be used to explain the present application, and should not be construed as limiting the present application.

[0046] As the level of intelligence of automobiles increases, the vehicle's sensor and actuator configurations are becoming more and more abundant. As the vehicle's perception ability becomes stronger and stronger, and as the computing power of intelligent vehicles increases, the vehicle can process more abundant environmental information, so that driving assistance functions can be realized based on multiple sensor information, such as adaptive cruise assistance, lane keeping, automatic parking, valet parking and other functions, which brings a lot of convenience to users, reduces driver fatigue and saves users' time.

[0047] Among them, valet parking, as one of the driving assistance functions, can help drivers automatically find parking spaces and realize automatic parking in parking scenarios such as garages, which can bring great convenience to users. Taking the garage as an example, due to the narrow roads in the garage, there are pedestrians, motor vehicles, non-motor vehicles and other targets with uncertain movement directions, and there are many intersections in the garage, making the working scene of valet parking extremely complex and changeable, especially the intersection scene in the garage. When crossing the intersection or turning right / left at the intersection, other target vehicles often enter the intersection at the same time. If the movement relationship between the vehicle and the target vehicle cannot be handled in time, it may cause the risk of collision between the vehicle and the target vehicle.

[0048] However, during the motion interaction process, if the vehicle chooses an overly conservative driving strategy, it will lead to a decrease in traffic efficiency and cause traffic congestion in the garage. If the vehicle chooses an overly aggressive control strategy, it may cause a collision risk due to excessive speed, or the vehicle and the target vehicle may enter the intersection at the same time, which will bring great challenges to the further planning of the vehicle.

[0049] To this end, the embodiment of the present application proposes a game-based solution for safe driving at intersections, which comprehensively considers human-machine interaction prompts, collision safety risks, and traffic efficiency, and conforms to people's driving habits. The following will describe the intersection driving method, device, vehicle, and storage medium provided by the embodiment of the present application in conjunction with the accompanying drawings.

[0050] Specifically, Figure 1 A schematic flow chart of a method for passing through an intersection for an autonomous driving vehicle provided in an embodiment of the present application.

[0051] like Figure 1 As shown, the intersection passing method of the autonomous driving vehicle includes the following steps:

[0052] In step S101, it is identified whether there is a target vehicle within a preset range of the target intersection.

[0053] The preset range of the target intersection may be a range preset by the user, for example Figure 2 As shown: a preset distance range in each direction of the intersection, such as a range of 10 meters, or a target is located within a preset time interval range at the intersection, such as a time interval range of 4 seconds, which is not specifically limited here.

[0054] It can be understood that the embodiments of the present application can be applied to intersections in various scenarios, such as intersections in underground garages, etc., without specific limitation; and the embodiments of the present application can identify target vehicles in various ways, such as sensors with real-time perception functions such as millimeter-wave radars and cameras, which are not specifically limited here.

[0055] Taking the garage application scenario as an example, and the vehicle is equipped with a millimeter-wave radar and a camera, the embodiment of the present application can use the millimeter-wave radar and the camera to jointly identify the target vehicle, thereby assisting the camera to perform more accurate target recognition in areas with poor visibility in the garage.

[0056] Specifically, if Figure 2 As shown, when the vehicle approaches an intersection, only targets within a certain distance range of the intersection are taken as collision risk targets, and an area of ​​interest is delineated when the vehicle passes through the intersection. The area of ​​interest defaults to an area 10 meters away from the intersection. When the vehicle approaches an intersection, if there are vehicles approaching within a preset distance in each direction of the intersection, such as within a range of 10 meters, then the vehicle can be determined to be a target vehicle.

[0057] Therefore, the embodiments of the present application can identify the target vehicle based on the distance from the intersection, and can also consider the motion state of different targets and determine the target vehicle based on the motion state. Specifically: when a moving target approaches the intersection, the speed of the moving target is used as a reference. When the target is within a preset time distance of the intersection, for example, within a 4-second time distance, the target is regarded as a target within the area of ​​interest when the vehicle passes through the intersection, and a collision risk assessment is performed on the target.

[0058] In an embodiment of the present application, before identifying whether a target vehicle exists within a preset range of the target intersection, it may include: obtaining environmental information and / or positioning information around the vehicle; identifying environmental information to obtain a first intersection identifier, and / or matching a second intersection identifier based on the positioning information and a preset map; and determining the target intersection based on the first intersection identifier and / or the second intersection identifier.

[0059] Among them, environmental information may include attributes of the target vehicle and ground signs, including vehicle type, vehicle position, vehicle speed, vehicle direction, road arrow type, arrow distance, etc. The embodiment of the present application can obtain environmental information around the vehicle through sensors such as cameras.

[0060] The positioning information may be obtained through one or more positioning technologies including GPS (Global Positioning System) technology, wireless communication technology (GSM / GPRS / CDMA), image processing technology, and GIS technology.

[0061] Among them, the preset map can be a high-precision map, etc., which can be selected according to actual conditions without specific limitation.

[0062] It can be understood that the embodiments of the present application can obtain static scene information based on high-precision maps and the positioning of the vehicle, including road boundaries, environmental characteristics, etc., and the dynamic information obtained includes whether there is an intersection in front of the current position of the vehicle, the longitudinal distance information between the intersection and the vehicle, etc., so that a variety of methods can be used to determine the intersection signs, so as to facilitate the determination of the target intersection, and effectively improve the feasibility and applicability of target intersection identification.

[0063] Specifically, the camera can output the type of road marking arrows and the horizontal and vertical distances of road marking arrows. Based on the positioning of the vehicle, the high-precision map outputs dynamic information within a certain distance range in front of the vehicle's current position, including whether there is an intersection in front of the vehicle's current position and the longitudinal distance between the intersection and the vehicle.

[0064] Based on garage construction standards and specifications, clear arrows are required to indicate the direction of the intersection ahead at intersections and corners inside the garage. Therefore, it can be indirectly judged that when the camera recognizes that the road ahead of the vehicle is covered by road markings, it is considered that the vehicle will pass through an intersection ahead.

[0065] In an embodiment of the present application, a target intersection can be determined based on a first intersection identifier and a second intersection identifier, including: calculating a first confidence level of the second intersection identifier based on positioning information, and calculating a second confidence level of the first intersection identifier based on environmental information; when the first confidence level is greater than a first confidence level threshold, determining the target intersection based on the second intersection identifier; when the first confidence level is less than or equal to the first confidence level threshold, and the second confidence level is greater than the second confidence level threshold, determining the target intersection based on the first intersection identifier, otherwise, outputting no intersection information.

[0066] Among them, the first confidence refers to the reliability of the second intersection sign calculated based on the positioning information, and the second confidence refers to the reliability of the first intersection sign calculated based on the environmental information. Among them, the first confidence threshold and the second confidence threshold can be set according to actual conditions, for example, both can be set to 70%, etc., and no specific limitation is made here.

[0067] It can be understood that in the embodiments of the present application, multiple information can be integrated to determine the target intersection, avoid using a single method that may lead to misidentification of the target intersection, and improve the accuracy of target intersection identification, thereby ensuring stable driving of the vehicle by improving the accuracy of the information and ensuring a stable source of information.

[0068] Specifically, when the current vehicle positioning is accurate and covered by a high-precision map, the static information output by the high-precision map can be used as a stable prior information to judge the environmental information. When the camera and the high-precision map can output the intersection information at the same time, such as Figure 3 As shown, in the embodiment of the present application, different input sources can be selected for intersection scene recognition, and the arbitration logic is as follows:

[0069] (1) When the current position confidence is high, the arrow mark and the intersection distance output by the high-precision map are used as the basis for intersection judgment, which means that when the first confidence is greater than the first confidence threshold, the target intersection is determined according to the second intersection mark;

[0070] (2) When the current position confidence is low and the arrow mark confidence output by the camera is high, the arrow type and arrow distance output by the camera are used as the basis for intersection determination, which means that when the first confidence is less than or equal to the first confidence threshold and the second confidence is greater than the second confidence threshold, the target intersection is determined according to the first intersection mark;

[0071] (3) When the positioning confidence is low and the confidence of the arrow mark output by the camera is low, in order to prevent scene misjudgment, the no intersection information is output.

[0072] In an embodiment of the present application, calculating the first confidence of the second intersection sign according to the positioning information may include: establishing a positioning covariance matrix according to the positioning information; and obtaining the first confidence by matching the variance value and the covariance value of the positioning covariance matrix.

[0073] Among them, the variance value refers to the measure of the degree of discreteness when measuring a random variable or a set of data, that is, the degree of deviation. The covariance value is used to measure the overall error of two variables, and the positioning covariance matrix is ​​established based on the positioning information to measure the intersection sign error.

[0074] It can be understood that in the embodiment of the present application, a positioning covariance matrix is ​​established based on the positioning information, and the first confidence level is obtained by matching the variance value and covariance value of the positioning covariance matrix, so that the credibility of the second intersection sign can be accurately determined, thereby improving the user experience.

[0075] Specifically, fusion positioning is based on sensor semantic information to assist vehicles in accurate positioning. The variance and covariance values ​​in the positioning covariance matrix can represent the accuracy of the current vehicle positioning. In the positioning covariance matrix, the diagonal x′→x, y′→y, and z′→z represent the variance in the three directions of longitude, latitude, and altitude. The smaller the variance value, the higher the positioning accuracy:

[0076] x′→xx′→yx′→z

[0077] y′→xy′→yy′→z

[0078] z′→xz′→yz′→z

[0079] By calibrating the positioning variance in the x, y, and z directions, it is determined that when the covariance is less than a specific calibration value, the current positioning result is more accurate, and the dynamic information of the high-precision map can be used as the basis for behavioral decision-making for the operation of the autonomous driving vehicle.

[0080] In the embodiment of the present application, calculating the second confidence level of the first intersection sign according to the environmental information may include: identifying the environmental information to obtain the actual environmental brightness; and obtaining the second confidence level according to the actual environmental brightness matching.

[0081] The actual environment brightness refers to the brightness of the actual environment obtained based on the environmental information in different actual environments. For example, when in a garage, the actual environment brightness recognized is generally darker; when passing an intersection on a sunny day, the actual environment brightness recognized is generally brighter.

[0082] It can be understood that in the embodiment of the present application, the credibility of the first intersection sign is determined quickly and accurately based on the ambient brightness, thereby improving the user experience.

[0083] In an embodiment of the present application, before identifying whether there is a target vehicle within a preset range of the target intersection, it may include: when the target intersection is identified, controlling the speed of the vehicle to be reduced to the target speed, and controlling the vehicle to perform a target reminder action.

[0084] Among them, the target vehicle speed can be set according to the actual situation, for example, it can be set to 3km / h or any speed value not higher than 3km / h, etc., and no specific limitation is made here; the target reminder actions may include honking, switching between high and low beams, and human-computer interaction reminders by the vehicle computer.

[0085] It is understandable that when passing through an intersection, in order to reduce the risk of collision when passing through the intersection, it is necessary to control the vehicle to pass through the intersection stably at a lower speed. The speed of the vehicle when it reaches the intersection can be set not higher than the target speed, such as 3km / h; at the same time, at a certain distance close to the intersection, such as about 5 meters, the vehicle is controlled to switch between high and low beams or automatically sound the horn to promptly alert surrounding target vehicles.

[0086] Specifically, when this vehicle passes through an intersection, it outputs the system response of the vehicle passing through the intersection based on environmental information and information of surrounding target vehicles, to ensure that the vehicle passes through the intersection safely and stably. When passing through an intersection, in order to ensure that it can attract the attention of other vehicles in time, it is necessary to switch between high and low beams or automatically sound the horn when approaching the intersection to remind target vehicles and pedestrians approaching the intersection. Similarly, when there is a driver in this vehicle, it is necessary to use human-computer interaction to remind the driver and passengers that the vehicle is about to approach the intersection, to attract the attention of the passengers in the car in time, and to ensure that when the vehicle encounters an unexpected collision risk when passing through the intersection, the driver can take over the vehicle in time.

[0087] In step S102, when the target vehicle is identified, the actual collision level is predicted according to the motion states of the host vehicle and the target vehicle, and the first expected deceleration of the host vehicle is matched with the second expected deceleration of the target vehicle according to the actual collision level.

[0088] Among them, the collision level can represent the probability of a collision between the vehicle and the target vehicle. The higher the collision level, the greater the corresponding probability. Therefore, in order to avoid a collision between the vehicle and the target vehicle, the corresponding expected deceleration is greater; the first expected deceleration refers to the deceleration caused by the target vehicle to the vehicle, and the second expected deceleration refers to the deceleration caused by the vehicle to the target vehicle.

[0089] It can be understood that, assuming that the target vehicle and this vehicle will enter the intersection at the same time, the actual collision level of the motion state of this vehicle and the target vehicle is relatively high. If this vehicle chooses to give way, it needs to decelerate at a specific deceleration to avoid a collision with the target vehicle. The deceleration is defined as the deceleration caused by the target vehicle to this vehicle, which is the first expected deceleration; assuming that the target vehicle and this vehicle will enter the intersection at the same time, the actual collision level of the motion state of the target vehicle and this vehicle is relatively high. If the target vehicle chooses to give way, it needs to decelerate at a specific deceleration to avoid a collision with the target vehicle (this vehicle). The deceleration is defined as the deceleration caused by this vehicle to the target vehicle, which is the second expected deceleration.

[0090] Specifically, if Figure 4 As shown in the figure, when the host vehicle and the target vehicle cross into the intersection, the concepts of deceleration caused by the host vehicle on the target and deceleration caused by the target on the host vehicle are introduced to evaluate the reference standard for vehicles to negotiate passing through the intersection when meeting at the intersection.

[0091] The deceleration caused by the target on the vehicle is defined as:

[0092] Assuming that the target vehicle and the own vehicle will enter the intersection at the same time, if the own vehicle chooses to give way, it needs to decelerate at a specific deceleration to avoid a collision with the target vehicle. The deceleration is defined as the deceleration caused by the target vehicle on the own vehicle.

[0093] Conversely, the deceleration caused by the vehicle on the target is defined as:

[0094] Assuming that the target vehicle and our vehicle will enter the intersection at the same time, if the target vehicle chooses to give way, it needs to decelerate at a specific deceleration to avoid a collision with the target vehicle. This deceleration is defined as the deceleration caused by our vehicle on the target vehicle.

[0095] In an embodiment of the present application, predicting the actual collision level based on the motion states of the vehicle and the target vehicle may include: predicting whether there is an overlapping area between the preset safety ranges of the vehicle and the target vehicle based on the motion states of the vehicle and the target vehicle; if there is an overlapping area, matching the actual collision level based on the area of ​​the overlapping area, wherein the area of ​​the overlapping area is positively correlated with the actual collision level, and a larger actual collision level indicates a higher collision risk; if there is no overlapping area, controlling the vehicle to maintain the current speed through the target intersection using the current driving strategy.

[0096] Among them, the preset safety range refers to the safe distance between the vehicle and the target vehicle, which can be set according to actual conditions, such as 0.5 meters, and is not specifically limited here.

[0097] It can be understood that the embodiments of the present application can quickly and accurately determine the degree of collision risk based on the degree of overlap between the safety range of the host vehicle and the target vehicle, so as to facilitate the determination of subsequent intersection traffic strategies.

[0098] Specifically, when predicting the actual collision level based on the motion state of the vehicle and the target vehicle, it is first necessary to expand the safety distance between the vehicle and the target vehicle. According to the size of a general car, the safety distance that needs to be expanded is set to 0.5 meters. When the BoundingBox (bounding box, including the expanded boundary) of the vehicle and the target vehicle have overlapping areas, it is considered that there is a risk of collision between the vehicle and the target vehicle, and the higher the area of ​​the overlapping area, the higher the collision risk. Therefore, when performing trajectory planning and speed planning, it is necessary to ensure that the BoundingBoxes of the two vehicles do not overlap.

[0099] In step S103, when the first expected deceleration is greater than the first deceleration threshold and the second expected deceleration is greater than the second deceleration threshold, the vehicle is controlled to accelerate or maintain the current speed to pass the target intersection according to the preset acceleration strategy or the current driving strategy; otherwise, the vehicle is controlled to decelerate and pass the target intersection according to the preset deceleration strategy.

[0100] The first deceleration threshold and the second deceleration threshold can be set according to actual conditions, for example, both can be set to -1.5 m / s^2, etc.

[0101] It can be understood that a reasonable intersection traffic strategy is determined by comprehensively considering the collision risk between the vehicle and the target vehicle. When the collision risk is low, when the first expected deceleration is greater than the first deceleration threshold and the second expected deceleration is greater than the second deceleration threshold, an active traffic strategy can be used to pass through the intersection to ensure intersection traffic efficiency. Otherwise, a conservative traffic strategy can be used to pass through the intersection to ensure intersection traffic safety, thereby taking into account both intersection traffic safety and efficiency and improving user experience.

[0102] For example, the embodiment of the present application calculates through collision detection that when the vehicle passes through the intersection with a deceleration of not less than -1.5m / s^2 (an acceptable deceleration), and the target vehicle passes through the intersection at a speed of not less than -1.5m / s^2, and the vehicles will not collide, the vehicle chooses to pass through the intersection normally at the current speed; when the deceleration caused by the target vehicle on the vehicle is less than -1.5m / s^2, the vehicle must choose a stop and yield strategy when approaching the intersection, and start driving after waiting for the target vehicle to pass the intersection; conversely, when the deceleration caused by the vehicle on the target vehicle is less than -1.5m / s^2, there is a risk that the target vehicle will not slow down to yield, and the vehicle must control the vehicle to slow down and stop at the intersection, and start driving after the target vehicle passes.

[0103] In the embodiment of the present application, the embodiment of the present application can also design a speed control strategy for the vehicle when passing through an intersection based on traffic rules, driving comfort, and collision analysis results. When passing through an intersection, the traffic rules for passing through an unprotected intersection need to be considered:

[0104] 1. When meeting at an intersection, vehicles on the right road have priority;

[0105] 2. The turning vehicle shall give priority to the executing motor vehicle;

[0106] 3. For vehicles traveling in opposite directions, right-turning vehicles should give way to left-turning vehicles.

[0107] When the trajectory of the own vehicle intersects with the trajectory of the target vehicle, it means that the own vehicle will collide with the target vehicle at some point in the future. At this time, it is assumed that the target vehicle has three states: accelerating to pass, maintaining the current speed to pass, and decelerating to give way. At the same time, the acceleration and deceleration boundaries allowed for the own vehicle when accelerating or decelerating to give way are limited. Through the optimization algorithm, the speed boundary that takes into account the intention of the target vehicle, the comfort of the own vehicle, and the collision risk is obtained to ensure that the vehicle can pass safely at this sequence of speeds.

[0108] According to the intersection passing method of an autonomous driving vehicle proposed in the embodiment of the present application, by considering the collision risk analysis between the vehicle and the target vehicle, the deceleration caused by the vehicle to the target vehicle, and the deceleration caused by the target vehicle to the vehicle, a safe driving strategy for the vehicle when passing through the intersection is finally obtained, which can not only ensure efficient traffic efficiency, but also ensure the comfort and safety of passengers. Thus, the problem that autonomous driving vehicles usually pass through intersections through conservative or radical strategies in related technologies, which cannot take into account the safety and efficiency of intersection passing, and the user experience is poor is solved.

[0109] In specific applications, the method of the embodiment of the present application can be applied to Figure 5 In the system architecture of the intersection driving scene shown, the system architecture includes four modules: perception, cognition, decision planning and control execution, and the intersection driving scene is simulated through the four modules. Among them, the perception module mainly relies on real-time perception sensors: cameras, millimeter-wave radars and high-precision maps with prior information. The cognitive module identifies the intersection scene based on the information perceived by the perception module; the decision planning module must output the system response of the vehicle passing through the intersection based on the environmental information and the surrounding target vehicle information when passing through the intersection to ensure that the vehicle passes through the intersection safely and stably; the control execution module is used to control the vehicle to perform corresponding actions according to the planning of the decision planning module.

[0110] based on Figure 5The system architecture of the intersection driving scenario shown in the figure is as follows:

[0111] S1: Integrated multi-source sensor recognition of intersection scenes;

[0112] The positioning covariance matrix ensures the accuracy of the positioning of intelligent driving vehicles in scenarios covered by high-precision maps. When the valet parking system is activated, the positioning variance of the positioning covariance matrix is ​​used as a way to arbitrate the intersection information recognized by the camera and the high-precision map. The map is a priori information source. When the positioning confidence is high, it can be used as a stable information source for judgment.

[0113] In scenarios covered by high-precision maps, when the positioning confidence meets the set confidence threshold, the intersection information of the high-precision map is used as the final information reference.

[0114] When the positioning confidence is lower than the set confidence threshold, the arrow mark and arrow mark distance recognized by the camera are used as the final information reference.

[0115] S2: Conduct intersection collision analysis to obtain the area of ​​interest for intersection driving and the speed boundary of the vehicle passing through the intersection;

[0116] To ensure driving safety, the minimum safety distance between the vehicle and the surrounding target vehicles in the garage intersection scene is expanded. Based on the vehicle's BoudingBox (bounding box), 0.5 meters around the vehicle's outer dimensions is set as the vehicle's safety boundary threshold. When performing collision detection, it is necessary to ensure that the BoudingBox between any two vehicles and the surrounding safety area have no intersection.

[0117] The garage intersection is an unprotected intersection. When the vehicle approaches the intersection, a certain area of ​​the intersection needs to be set as the area of ​​interest during driving at the intersection. The targets of interest when the vehicle passes through the intersection are screened out based on the boundaries of the area of ​​interest. Through collision analysis of the targets of interest, the speed boundary of the vehicle passing through the intersection is obtained.

[0118] S3: Based on traffic rules, driving comfort, and collision analysis results, design a speed control strategy for the vehicle when passing through an intersection;

[0119] When crossing an intersection, you need to consider the traffic rules for crossing an unprotected intersection:

[0120] 1. When meeting at an intersection, vehicles on the right road have priority;

[0121] 2. Motor vehicles turning should give priority to motor vehicles executing the traffic rules;

[0122] 3. Vehicles traveling in opposite directions, right-turning vehicles should give way to left-turning vehicles;

[0123] When the trajectory of the own vehicle intersects with the trajectory of the target vehicle, it means that the own vehicle will collide with the target vehicle at some point in the future. At this time, it is assumed that the target vehicle has three states: accelerating to pass, maintaining the current speed to pass, and decelerating to give way. At the same time, the acceleration and deceleration boundaries allowed for the own vehicle when accelerating or decelerating to give way are limited. Through the optimization algorithm, the speed boundary that takes into account the intention of the target vehicle, the comfort of the own vehicle, and the collision risk is obtained to ensure that the vehicle can pass safely at this sequence of speeds.

[0124] S4: Design human-machine interaction strategies for intersection driving based on safety and system intelligence;

[0125] When passing through an intersection, in order to ensure that you can attract the attention of other vehicles in time, you need to switch between high and low beams when approaching the intersection to alert target vehicles and pedestrians approaching the intersection.

[0126] Similarly, when there is a driver in the vehicle, it is necessary to use human-computer interaction to remind the driver and passengers that the vehicle is approaching an intersection, to promptly draw the attention of the passengers, and to ensure that the driver can take over the vehicle in time when the vehicle encounters an unexpected collision risk when passing through the intersection.

[0127] In summary, when the intelligent valet parking system is driving in the garage and the vehicle is driving at the intersection, in order to ensure stable and safe passage through the intersection, information interaction prompts can be given to the surrounding pedestrians and targets. When the vehicle and the surrounding target vehicles pass through the intersection at the same time, the vehicle can judge the driving intention of the surrounding target vehicles, accurately estimate the driving trajectory of the vehicle and the surrounding target vehicles, plan a reasonable driving trajectory and driving speed, and pass through the intersection by playing games with the surrounding targets. Therefore, the embodiment of the present application can consider the collision risk analysis between the vehicle and the target vehicle, the deceleration caused by the vehicle to the target vehicle, and the deceleration caused by the target vehicle to the vehicle, and finally obtain a safe driving strategy for the vehicle when passing through the intersection, which can not only ensure efficient traffic efficiency, but also ensure the comfort and safety of passengers.

[0128] Next, an intersection passage device for an autonomous driving vehicle proposed according to an embodiment of the present application will be described with reference to the accompanying drawings.

[0129] Figure 6 It is a block diagram of an intersection passage device for an autonomous driving vehicle according to an embodiment of the present application.

[0130] like Figure 6 As shown, the intersection passage device 10 of the autonomous driving vehicle may include: a recognition module 100, a matching module 200 and a control module 300.

[0131] Among them, the identification module 100 is used to identify whether there is a target vehicle within a preset range of the target intersection; the matching module 200 is used to predict the actual collision level according to the motion state of the vehicle and the target vehicle when the target vehicle is identified, and match the first expected deceleration of the vehicle with the second expected deceleration of the target vehicle according to the actual collision level; the control module 300 is used to control the vehicle to accelerate or maintain the current speed to pass the target intersection according to the preset acceleration strategy or the current driving strategy when the first expected deceleration is greater than the first deceleration threshold and the second expected deceleration is greater than the second deceleration threshold, otherwise control the vehicle to decelerate and pass the target intersection according to the preset deceleration strategy.

[0132] In an embodiment of the present application, the matching module 200 can also be used to: predict whether there is an overlapping area in the preset safety range of the vehicle and the target vehicle according to the motion state of the vehicle and the target vehicle; if there is an overlapping area, match the actual collision level according to the area of ​​the overlapping area, wherein the area of ​​the overlapping area is positively correlated with the actual collision level, and the larger the actual collision level, the higher the collision risk; if there is no overlapping area, control the vehicle to maintain the current speed through the target intersection according to the current driving strategy.

[0133] In the embodiment of the present application, the device 10 of the embodiment of the present application may include: a determination module. The determination module is used to obtain environmental information and / or positioning information around the vehicle before identifying whether there is a target vehicle within a preset range of the target intersection; identify the environmental information to obtain a first intersection identifier, and / or match a second intersection identifier according to the positioning information and a preset map; and determine the target intersection according to the first intersection identifier and / or the second intersection identifier.

[0134] In an embodiment of the present application, the determination module is further used to: calculate a first confidence level of the second intersection sign based on the positioning information, and calculate a second confidence level of the first intersection sign based on the environmental information; when the first confidence level is greater than a first confidence level threshold, determine the target intersection based on the second intersection sign; when the first confidence level is less than or equal to the first confidence level threshold, and the second confidence level is greater than the second confidence level threshold, determine the target intersection based on the first intersection sign, otherwise, output no intersection information.

[0135] In an embodiment of the present application, the determination module is further used to: establish a positioning covariance matrix based on the positioning information; obtain a first confidence level by matching the variance value and covariance value of the positioning covariance matrix; the determination module is further used to: identify environmental information to obtain the actual ambient brightness; and obtain a second confidence level by matching the actual ambient brightness.

[0136] In an embodiment of the present application, the control module 300 can be further used to: before identifying whether there is a target vehicle within a preset range of the target intersection, when the target intersection is identified, control the vehicle speed to be reduced to the target speed, and control the vehicle to perform a target reminder action.

[0137] It should be noted that the aforementioned explanation of the embodiment of the method for passing through an intersection for an autonomous driving vehicle is also applicable to the intersection passing device for the autonomous driving vehicle of this embodiment, and will not be repeated here.

[0138] According to the intersection passing device of the autonomous driving vehicle proposed in the embodiment of the present application, the target vehicle is identified, the collision risk analysis between the vehicle and the target vehicle is matched, the deceleration caused by the vehicle to the target vehicle, and the deceleration caused by the target vehicle to the vehicle are matched, and finally the vehicle is controlled to pass through the intersection, which can ensure efficient passage efficiency and the comfort and safety of passengers. Thus, the problem that autonomous driving vehicles in related technologies usually pass through intersections through conservative or radical strategies, which cannot take into account the safety and efficiency of intersection passage, and the user experience is poor.

[0139] Figure 7 A schematic diagram of the structure of a vehicle provided in an embodiment of the present application. The vehicle may include:

[0140] A memory 701 , a processor 702 , and a computer program stored in the memory 701 and executable on the processor 702 .

[0141] When the processor 702 executes the program, the intersection passage method of the autonomous driving vehicle provided in the above embodiment is implemented.

[0142] Furthermore, the vehicle may also include:

[0143] The communication interface 703 is used for communication between the memory 701 and the processor 702 .

[0144] The memory 701 is used to store computer programs that can be executed on the processor 702 .

[0145] The memory 701 may include a high-speed RAM (Random Access Memory) memory, and may also include a non-volatile memory, such as at least one disk memory.

[0146] If the memory 701, the processor 702 and the communication interface 703 are implemented independently, the communication interface 703, the memory 701 and the processor 702 can be connected to each other through a bus and communicate with each other. The bus can be an ISA (Industry Standard Architecture) bus, a PCI (Peripheral Component Interconnect) bus or an EISA (Extended Industry Standard Architecture) bus, etc. The bus can be divided into an address bus, a data bus, a control bus, etc. For ease of representation, Figure 7 Only one thick line is used in the diagram, but this does not mean that there is only one bus or only one type of bus.

[0147] Optionally, in a specific implementation, if the memory 701, the processor 702 and the communication interface 703 are integrated on a chip, the memory 701, the processor 702 and the communication interface 703 can communicate with each other through an internal interface.

[0148] The processor 702 may be a CPU (Central Processing Unit), or an ASIC (Application Specific Integrated Circuit), or one or more integrated circuits configured to implement the embodiments of the present application.

[0149] An embodiment of the present application also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the above-mentioned intersection passage method for an autonomous driving vehicle.

[0150] In the description of this specification, the description with reference to the terms "one embodiment", "some embodiments", "example", "specific example", or "some examples" etc. means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present application. In this specification, the schematic representations of the above terms are not necessarily directed to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described may be combined in any one or N embodiments or examples in a suitable manner. In addition, those skilled in the art may combine and combine the different embodiments or examples described in this specification and the features of the different embodiments or examples, without contradiction.

[0151] In addition, the terms "first" and "second" are used for descriptive purposes only and should not be understood as indicating or implying relative importance or implicitly indicating the number of technical features indicated. Therefore, a feature defined as "first" or "second" may explicitly or implicitly include at least one of the features. In the description of this application, "N" means at least two, such as two, three, etc., unless otherwise clearly and specifically defined.

[0152] Any process or method description in a flowchart or otherwise described herein may be understood to represent a module, fragment or portion of code comprising one or more executable instructions for implementing the steps of a custom logical function or process, and the scope of the preferred embodiments of the present application includes alternative implementations in which functions may not be performed in the order shown or discussed, including performing functions in a substantially simultaneous manner or in reverse order depending on the functions involved, which should be understood by technicians in the technical field to which the embodiments of the present application belong.

[0153] It should be understood that the various parts of the present application can be implemented by hardware, software, firmware or a combination thereof. In the above-mentioned embodiment, the N steps or methods can be implemented by software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented by hardware, as in another embodiment, it can be implemented by any one of the following technologies known in the art or their combination: a discrete logic circuit having a logic gate circuit for implementing a logic function for a data signal, a dedicated integrated circuit having a suitable combination of logic gate circuits, a programmable gate array, a field programmable gate array, etc.

[0154] A person skilled in the art may understand that all or part of the steps in the method for implementing the above-mentioned embodiment may be completed by instructing related hardware through a program, and the program may be stored in a computer-readable storage medium, which, when executed, includes one or a combination of the steps of the method embodiment.

[0155] Although the embodiments of the present application have been shown and described above, it can be understood that the above embodiments are exemplary and cannot be understood as limitations on the present application. Ordinary technicians in this field can change, modify, replace and modify the above embodiments within the scope of the present application.

Claims

1. A method for an autonomous driving vehicle to pass through an intersection, characterized in that: The following steps are involved: Identify whether there is a target vehicle within a preset range of the target intersection; When the target vehicle is identified, an actual collision level is predicted according to the motion states of the host vehicle and the target vehicle, and a first expected deceleration of the host vehicle is matched with a second expected deceleration of the target vehicle according to the actual collision level; When the first expected deceleration is greater than the first deceleration threshold and the second expected deceleration is greater than the second deceleration threshold, the vehicle is controlled to accelerate or maintain the current speed through the target intersection according to the preset acceleration strategy or the current driving strategy; otherwise, the vehicle is controlled to decelerate and pass through the target intersection according to the preset deceleration strategy.

2. The method according to claim 1, characterized in that The predicting of the actual collision level according to the motion state of the host vehicle and the target vehicle includes: Predicting whether there is an overlapping area between the preset safety ranges of the host vehicle and the target vehicle according to the motion states of the host vehicle and the target vehicle; If there is an overlapping area, matching the actual collision level according to the area of ​​the overlapping area, wherein the area of ​​the overlapping area is positively correlated with the actual collision level, and a larger actual collision level indicates a higher collision risk; If the overlapping area does not exist, the vehicle is controlled to maintain the current speed through the target intersection according to the current driving strategy.

3. The method according to claim 1, characterized in that Before identifying whether there is a target vehicle within a preset range of the target intersection, the method includes: Obtaining environmental information and / or positioning information around the vehicle; Identify the environmental information to obtain a first intersection identifier, and / or match a second intersection identifier according to the positioning information and a preset map; The target intersection is determined according to the first intersection identifier and / or the second intersection identifier.

4. The method according to claim 3, characterized in that Determining the target intersection according to the first intersection identifier and the second intersection identifier includes: Calculating a first confidence level of the second intersection marker according to the positioning information, and calculating a second confidence level of the first intersection marker according to the environmental information; When the first confidence level is greater than a first confidence level threshold, determining the target intersection according to the second intersection identifier; When the first confidence is less than or equal to the first confidence threshold and the second confidence is greater than the second confidence threshold, the target intersection is determined according to the first intersection identifier; otherwise, no intersection information is output.

5. The method according to claim 4, characterized in that The calculating a first confidence level of the second intersection identifier according to the positioning information includes: Establishing a positioning covariance matrix according to the positioning information; The first confidence level is obtained by matching the variance value and the covariance value of the positioning covariance matrix.

6. The method according to claim 4, characterized in that Calculating a second confidence level of the first intersection sign according to the environmental information includes: Identify the environmental information to obtain actual environmental brightness; The second confidence level is obtained according to the actual environment brightness matching.

7. The method according to any one of claims 1 to 6, characterized in that: Before identifying whether there is a target vehicle within a preset range of the target intersection, the method includes: When the target intersection is identified, the vehicle speed of the host vehicle is controlled to be reduced to the target speed, and the host vehicle is controlled to perform a target reminder action.

8. An intersection passing device for an autonomous driving vehicle, characterized in that: include: An identification module, used to identify whether a target vehicle exists within a preset range of a target intersection; a matching module, configured to predict an actual collision level according to the motion states of the host vehicle and the target vehicle when the target vehicle is identified, and to match a first expected deceleration of the host vehicle with a second expected deceleration of the target vehicle according to the actual collision level; A control module is used to control the vehicle to accelerate or maintain the current speed to pass the target intersection according to a preset acceleration strategy or a current driving strategy when the first expected deceleration is greater than a first deceleration threshold and the second expected deceleration is greater than a second deceleration threshold; otherwise, the vehicle is controlled to decelerate and pass the target intersection according to a preset deceleration strategy.

9. An autonomous driving vehicle, characterized in that: include: A memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the intersection passage method for an autonomous driving vehicle as described in any one of claims 1 to 7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: The program is executed by a processor to implement the intersection passage method of an autonomous driving vehicle as described in any one of claims 1-7.

Images