Method, device and autonomous vehicle for determining an obstacle causing a predetermined behavior

Abstract

The present disclosure provides a method, device and automatic driving vehicle for determining an obstacle causing a predetermined behavior, relates to the field of artificial intelligence, and particularly to the field of automatic driving. The specific implementation scheme is: determining a target time according to an abnormal time when the vehicle has the predetermined behavior; determining N obstacles from a plurality of candidate obstacles according to the target time, position information of the vehicle and position information of each of the plurality of candidate obstacles; N is an integer greater than or equal to 1; determining whether a speed of each of the N obstacles appears a false detection at the target time, obtaining at least one first obstacle appearing the speed false detection; and determining a target obstacle causing the vehicle to have the predetermined behavior from the at least one first obstacle according to relevant position information of the at least one first obstacle related to the target time and a distance threshold.

Description

TECHNICAL FIELD

[0001] The present disclosure relates to the technical field of artificial intelligence, in particular to the field of automatic driving, and more particularly, the present disclosure provides a method, device, electronic device, storage medium, computer program product and automatic driving vehicle for determining an obstacle causing a predetermined behavior. BACKGROUND

[0002] In an automatic driving scenario, a perception model may sometimes incorrectly judge the motion state of an obstacle, for example, judge that a stationary obstacle has some speed, causing an automatic driving vehicle to take emergency braking or lane changing measures based on the incorrect judgment result, affecting the user riding experience. SUMMARY

[0003] The present disclosure provides a method, device, electronic device, storage medium, computer program product and automatic driving vehicle for determining an obstacle causing a predetermined behavior.

[0004] According to an aspect of the present disclosure, a method for determining an obstacle causing a predetermined behavior is provided, including: determining a target time according to an abnormal time when a vehicle takes a predetermined behavior; determining N obstacles from a plurality of candidate obstacles according to the target time, position information of the vehicle and respective position information of the plurality of candidate obstacles; N is an integer greater than or equal to 1; determining whether a speed of each of the N obstacles appears to be misdetected at the target time to obtain at least one first obstacle appearing to be misdetected in speed; and determining a target obstacle causing the vehicle to take the predetermined behavior from the at least one first obstacle according to relevant position information of the at least one first obstacle related to the target time and a distance threshold.

[0005] According to another aspect of the present disclosure, a device for determining an obstacle causing a predetermined behavior is provided, including: a first determining module, a second determining module, a third determining module and a fourth determining module. The first determining module is configured to determine a target time according to an abnormal time when a vehicle takes a predetermined behavior. The second determining module is configured to determine N obstacles from a plurality of candidate obstacles according to the target time, position information of the vehicle and respective position information of the plurality of candidate obstacles; N is an integer greater than or equal to 1. The third determining module is configured to determine whether a speed of each of the N obstacles appears to be misdetected at the target time to obtain at least one first obstacle appearing to be misdetected in speed. The fourth determining module is configured to determine a target obstacle causing the vehicle to take the predetermined behavior from the at least one first obstacle according to relevant position information of the at least one first obstacle related to the target time and a distance threshold.

[0006] According to another aspect of the present disclosure, an electronic device is provided, comprising: at least one processor; and a memory connected with the at least one processor in communication; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to perform the method provided by the present disclosure.

[0007] According to another aspect of the present disclosure, a non-transitory computer readable storage medium storing computer instructions is provided, wherein the computer instructions are used to enable a computer to perform the method provided by the present disclosure.

[0008] According to another aspect of the present disclosure, a computer program product is provided, comprising a computer program, which, when executed by a processor, implements the method provided by the present disclosure.

[0009] According to another aspect of the present disclosure, an autonomous vehicle is provided, comprising the electronic device described above.

[0010] It should be understood that the content described in this part is not intended to identify key or important features of the embodiments of the present disclosure, nor to limit the scope of the present disclosure. Other features of the present disclosure will become apparent from the following description. BRIEF DESCRIPTION OF DRAWINGS

[0011] The accompanying drawings are used to better understand the present scheme, and do not limit the present disclosure. Among them:

[0012] Figure 1 is an application scenario diagram of the method and device for determining obstacles according to an embodiment of the present disclosure;

[0013] Figure 2 is a schematic flowchart of the method for determining obstacles according to an embodiment of the present disclosure;

[0014] Figure 3 is a schematic diagram of the relative position relationship between the vehicle body and the obstacle according to an embodiment of the present disclosure;

[0015] Figure 4 is a schematic flowchart of the method for determining obstacles according to another embodiment of the present disclosure;

[0016] Figure 5A is a schematic principle diagram of the method for determining obstacles according to an embodiment of the present disclosure;

[0017] Figure 5B is a schematic principle diagram of the method for determining obstacles according to an embodiment of the present disclosure;

[0018] Figure 6 is a schematic structural block diagram of the device for determining obstacles according to an embodiment of the present disclosure; and

[0019] Figure 7 is a structural block diagram of an electronic device for implementing a method of determining an obstacle according to an embodiment of the disclosure. DETAILED DESCRIPTION

[0020] Exemplary embodiments of the disclosure are described herein with reference to the accompanying drawings, which are presented for the purpose of illustration and description. It is to be understood that the embodiments described herein are exemplary only, and various changes and modifications can be made thereto without departing from the scope and spirit of the disclosure. Also, for the purpose of clarity and the brevity of description, the description below omits descriptions of well-known functions and structures.

[0021] Figure 1 is a schematic diagram of an application scenario of a method and apparatus for determining an obstacle according to an embodiment of the disclosure.

[0022] It should be noted that Figure 1 The system architecture shown is merely an example of a system architecture to which embodiments of the disclosure can be applied, to help those skilled in the art understand the technical content of the disclosure, but does not mean that embodiments of the disclosure cannot be used in other devices, systems, environments or scenarios.

[0023] As Figure 1 shown, the system architecture 100 according to this embodiment can include sensors 101, 102, 103, a network 120, a server 130 and a road side unit (RSU) 140. The network 120 is a medium for providing a communication link between the sensors 101, 102, 103 and the server 130. The network 120 can include various connection types, such as wired and / or wireless communication links, etc.

[0024] The sensors 101, 102, 103 can interact with the server 130 through the network 120 to receive or send messages, etc.

[0025] The sensors 101, 102, 103 can be functional elements integrated on the vehicle 110, such as infrared sensors, ultrasonic sensors, millimeter wave radars, image acquisition devices, laser radars, inertial measurement units, etc. The sensors 101, 102, 103 can be used to acquire state data of perceived objects (such as pedestrians, vehicles, obstacles, etc.) and surrounding road data around the vehicle 110.

[0026] The vehicle 110 can communicate with the road side unit 140, receive information from the road side unit 140, or send information to the road side unit.

[0027] The server 130 can be disposed at a remote end capable of establishing communication with the vehicle terminal, and can be implemented as a distributed server cluster composed of multiple servers, or as a single server.

[0028] The server 130 can be a server providing various services. For example, a map application, a data processing application, and the like can be installed on the server 130. Taking the server 130 running the data processing application as an example: receiving data such as vehicle driving decisions, speeds of obstacles, and the like transmitted from the sensors 101, 102, and 103 through the network 120, and taking the data as to-be-processed data. And processing the to-be-processed data to obtain a target obstacle causing the vehicle to make an emergency stop.

[0029] It should be noted that the method for determining an obstacle causing a predetermined behavior provided in the embodiments of the present disclosure can be generally executed by the server 130. Accordingly, the apparatus for determining an obstacle causing a predetermined behavior provided in the embodiments of the present disclosure can also be arranged in the server 130.

[0030] It can be understood that, Figure 1 The number of sensors, networks, and servers in the above-mentioned embodiments is only illustrative. According to the needs of implementation, there can be any number of sensors, networks, and servers.

[0031] Figure 2 is a schematic flowchart of the method for determining an obstacle according to the embodiments of the present disclosure.

[0032] As shown in Figure 2 , the method 200 for determining an obstacle causing a predetermined behavior can include operation S210 to operation S240.

[0033] At operation S210, a target time is determined according to an abnormal time when the vehicle makes a predetermined behavior.

[0034] For example, the predetermined behavior can include behaviors such as emergency stop, lane change, and the like.

[0035] For example, the vehicle makes a predetermined behavior at an abnormal time t s , a time before the abnormal time t s may be determined as the target time.

[0036] In one example, a time before the abnormal time t s and separated from the abnormal time t s by a predetermined duration can be determined as the target time.

[0037] In another example, it is assumed that the reaction time of the vehicle from perceiving danger to making a predetermined behavior is greater than t1 and less than t2, so it can be determined that the reaction period (t s -t2, t s -t1) of the vehicle, at least one target time can be determined from the reaction period (t s -t2, t s -t1).

[0038] In operation S220, N obstacles are determined from the plurality of candidate obstacles according to the target time, the position information of the vehicle, and the position information of each of the plurality of candidate obstacles; N is an integer greater than or equal to 1.

[0039] For example, the candidate obstacle can represent an obstacle detected by the vehicle around itself at the target time.

[0040] For example, the position of the vehicle at the target time and the position of each candidate obstacle at the target time can be determined according to the information perceived by the vehicle. Next, the candidate obstacle closer to the vehicle at the target time can be added to the N obstacles, or the obstacle in front of the vehicle at the target time can be added to the N obstacles, and the front can include the front and the side front.

[0041] In operation S230, it is determined whether the speed of each of the N obstacles is misdetected at the target time, and at least one first obstacle with speed misdetected is obtained.

[0042] For example, the speed misdetected can include that the speed of the obstacle in a stationary state is detected.

[0043] For example, the speed of the obstacle at the target time collected by other equipment can be obtained, and the other equipment can be a road testing device or other autonomous vehicle. In the case that the speed of a certain obstacle at the target time detected by the other equipment is 0, the other equipment is recorded, and in the case that the number of recorded other equipment is greater than a number threshold, it is determined that the speed of the obstacle is misdetected at the target time.

[0044] It can be understood that the N obstacles can include the first obstacle with speed misdetected, and also include a second obstacle without speed misdetected, and for the second obstacle, the processing flow of the second obstacle can be ended.

[0045] In operation S240, a target obstacle causing the vehicle to have a predetermined behavior is determined from the at least one first obstacle according to the relevant position information of each of the at least one first obstacle related to the target time and a distance threshold.

[0046] For example, the relevant position information includes position point information of the obstacle at the target time, and the actual distance between the two obstacles and the vehicle can be determined according to the position point information and the position of the vehicle at the target time, and the first obstacle with the actual distance less than or equal to the distance threshold is determined as the target obstacle.

[0047] According to the technical scheme provided by the embodiment of the present disclosure, the target moment with a large probability of speed misjudgment of the vehicle is determined first when the vehicle performs the predetermined behavior abnormally. Then, N suspicious obstacles are determined from the candidate obstacles appearing around the vehicle at the target moment. Next, it is determined whether the speed of the N obstacles is misjudged, and based on the obstacle whose speed is misjudged, the target obstacle that is more likely to cause the vehicle to perform the predetermined behavior such as sudden braking or lane changing is determined.

[0048] After obtaining the target obstacle, the type and other information of the target obstacle can be labeled, and then the labeled information is used to optimize the automatic driving model, thereby improving the accuracy of the automatic driving model in detecting the speed of the obstacle and alleviating the problem of sudden braking or lane changing of the vehicle due to misjudgment of the speed of the obstacle.

[0049] Figure 3 is a schematic diagram of the relative position relationship between the vehicle body and the obstacle according to the embodiment of the present disclosure.

[0050] As shown in Figure 3 , the N obstacles can be determined from the plurality of candidate obstacles according to the position of the vehicle at the target moment and the position of each candidate obstacle at the target moment.

[0051] In the process of determining the position of the vehicle, the vehicle can be simplified as a vehicle point, and the position of the vehicle is represented by the coordinates of the vehicle point. For example, a vehicle body coordinate system can be established, and the position of the vehicle is represented by the origin of the vehicle body coordinate system. It can be understood that the vehicle body coordinate system includes x-axis and y-axis perpendicular to each other, wherein the x-axis can extend along the vehicle width direction, and the y-axis can extend along the vehicle length direction.

[0052] In the process of determining the position of the obstacle, the candidate obstacle can be simplified as an obstacle point, and the position of the obstacle is represented by the coordinates of the obstacle point. For example, the position of the obstacle can be represented by the center position of the candidate obstacle.

[0053] In an example, it can be determined whether the position of the candidate obstacle and the position of the vehicle at the target moment are less than or equal to a first distance threshold first. If yes, the candidate obstacle can be added to the N obstacles. If no, it indicates that the distance between the candidate obstacle and the vehicle at the target moment is far, and the candidate obstacle is not easy to cause the vehicle to perform sudden braking, so the candidate obstacle can not be added to the N obstacles. The first distance threshold can be 200 meters, for example.

[0054] This example determines whether to add the candidate obstacle to the N obstacles according to the distance between the candidate obstacle and the vehicle, thereby filtering the obstacles that do not cause the vehicle to perform sudden braking, thereby reducing the amount of subsequent data processing, improving processing efficiency, and also improving the accuracy of the target obstacle.

[0055] In another example, the relative position relationship between the vehicle and the candidate obstacle at the target time can be determined according to the position of the vehicle at the target time and the position of each candidate obstacle at the target time, in combination with the heading direction of the vehicle, and the candidate obstacle whose relative position relationship meets a predetermined relative position relationship is added to the N obstacles.

[0056] The predetermined position relationship can include that the candidate obstacle is in front of the vehicle, for example, in the vehicle body coordinate system, the obstacle whose coordinate along the vehicle length direction (i.e., the y direction) is greater than 0 can be determined as a front obstacle. In addition, the front can include at least one of the front and the side front.

[0057] For example, when the coordinate of the obstacle along the vehicle length direction (i.e., the y direction) is greater than 0, and the absolute value of the coordinate along the vehicle width direction (i.e., the x direction) is greater than or equal to a threshold L s , it can be determined that the obstacle is located in the side front of the vehicle, for example Figure 3 The obstacle 302 is located in the side front of the vehicle.

[0058] For example, when the coordinate of the obstacle along the vehicle length direction (i.e., the y direction) is greater than 0, and the absolute value of the coordinate along the vehicle width direction (i.e., the x direction) is less than a threshold L s , it can be determined that the obstacle is located in the front of the vehicle, for example, the obstacle 301 is located in the front of the vehicle.

[0059] For example, the threshold L s may be calculated by the following formula:

[0060] L s = (l 车宽 +l 障碍物宽度 ) / 2+l 阈值

[0061] The present example determines whether to add the candidate obstacle to the N obstacles according to the relative position relationship between the candidate obstacle and the vehicle, which can filter out obstacles that will not cause the vehicle to brake suddenly, for example, filter out obstacles located in the rear of the vehicle, thereby reducing the amount of subsequent data processing, improving processing efficiency, and also improving the accuracy of the target obstacle.

[0062] According to another embodiment of the present disclosure, the method for determining whether the speed of each of the N obstacles is misdetected at the target time can include the following operations: determining whether the speed of the obstacle is misdetected at the target time according to the speed and / or position change information of the obstacle within a target period, wherein the starting time of the target period is before the target time, and the ending time of the target period is after the target time.

[0063] In one example, the target time period includes a first time period, a second time period, and a third time period, wherein the first time period is before the target time, for example, the first time period is (t s -T, t s -t2). The second time period is between the end time of the first time period and the abnormal time, the second time period includes the target time, for example, the second time period is (t s -t2, t s -t1). The third time period is after the abnormal time, for example, the third time period is (t s , t s +T). The speed of the obstacle can be determined to be false positive in a case that the speed of the obstacle satisfies a first condition, wherein the first condition can include: being static in the first time period, being non-static in the second time period, and being static in the third time period.

[0064] In another example, the target time period further includes a first time and a second time, wherein the first time is before the target time, for example, the first time is t s -T time. The second time is after the target time, for example, the second time is t s +T time. For the same obstacle, the speed of the obstacle can be determined to be false positive in a case that the position change information of the obstacle satisfies a second condition, wherein the second condition includes that the distance between the position of the obstacle at the first time and the position of the obstacle at the second time is less than a moving threshold, and the distance threshold can be 0.2 meters.

[0065] In another example, the speed of the obstacle can be determined to be false positive in a case that the speed of the obstacle satisfies the first condition and the position change information of the obstacle satisfies the second condition. It can be understood that according to the speed of the obstacle and the position change information of the obstacle, whether the speed of the obstacle is false positive can be accurately determined.

[0066] Figure 4 is a schematic flow chart of a method for determining an obstacle according to another embodiment of the disclosure.

[0067] As Figure 4 shown, the method 400 for determining an obstacle causing a predetermined behavior can include operations S410-S470, wherein operations S410-S430 can refer to S210-S230 above, and will not be described herein. S450-S470 can be executed before operation S440.

[0068] In operation S450, according to the planned trajectory of the vehicle at the reference time and the estimated trajectory of the obstacle at the target time, it is determined whether the vehicle and the obstacle exist a collision risk.

[0069] For example, the reference time can be the same as the target time. The reference time can also be different from the target time, for example, the reference time is t s at time t2.

[0070] For example, the estimated trajectory can represent a trajectory of the obstacle determined by the automatic driving algorithm according to the speed of the detected obstacle after the target time.

[0071] For example, it can be determined first whether the planned trajectory and the estimated trajectory intersect. If they intersect, it can be determined that the vehicle and the obstacle have a collision risk, and the position of intersection is the estimated collision position. If they do not intersect, the plurality of trajectory points in the estimated trajectory can be respectively extended to form a bounding box corresponding to each trajectory point.

[0072] Next, it can be determined whether the edge line segment of the bounding box has a collision risk with the planned trajectory of the vehicle.

[0073] For example, if the edge line segment of the bounding box intersects with the planned trajectory of the vehicle, it can be determined that the vehicle and the obstacle have a collision risk, and the position of intersection can be determined as the estimated collision position.

[0074] For example, if the edge line segment of the bounding box does not intersect with the planned trajectory of the vehicle, but the distance between the vertex of the bounding box and the planned trajectory of the vehicle is less than the second distance threshold, it can be determined that the vehicle and the obstacle have a collision risk. For example, the sum of half of the width of the vehicle and the distance allowance can be determined as the second distance threshold. For example, the point in the planned trajectory of the vehicle closest to the vertex can be determined as the estimated collision position.

[0075] For example, if the edge line segment of the bounding box does not intersect with the planned trajectory of the vehicle, and the distance between the vertex of the bounding box and the planned trajectory of the vehicle is greater than or equal to the second distance threshold, it can be determined that the vehicle and the obstacle do not have a collision risk.

[0076] In operation S460, in the case where it is determined that there is a collision risk, a collision distance corresponding to the obstacle is determined according to the estimated collision position and the position of the vehicle at the reference time.

[0077] For example, the straight-line distance between the position of the vehicle at the reference time and the estimated collision position can be determined as the collision distance corresponding to the obstacle.

[0078] For example, the length of the sub-trajectory in the planned trajectory between the position of the vehicle at the reference time and the estimated collision position can be determined as the collision distance corresponding to the obstacle.

[0079] It should be noted that for the obstacle that does not have a collision risk with the vehicle, the collision distance of the obstacle can be recorded as a predetermined value, such as infinity, null, etc.

[0080] It can be understood that the collision distance is related to the distance that the vehicle needs to move to collide with the obstacle, and the collision distance can evaluate the risk of causing the vehicle to suddenly stop due to the obstacle, and thus the obstacle that has a greater probability of causing the vehicle to suddenly stop can be determined based on the collision distance.

[0081] In operation S470, the distance threshold is determined.

[0082] In an example, the distance threshold can be set in advance, for example, the distance threshold is set to 10 meters.

[0083] In another example, the collision distance threshold can be determined according to the minimum value of the collision distance of the second obstacle, for example, the minimum value of the plurality of collision distances corresponding to the plurality of second obstacles is determined as the collision distance threshold.

[0084] It should be noted that for the first obstacle obs1 and obs2 whose speed is misdetected and the second obstacle obs3 whose speed is not misdetected, if the collision distance of the first obstacle obs1 is greater than the collision distance of the second obstacle obs3, and if the collision distance of the first obstacle obs2 is less than the collision distance of the second obstacle obs3. Since the speed of the second obstacle obs3 is not misdetected, the vehicle is not likely to suddenly stop based on the speed of the second obstacle obs3. Therefore, even if the speed of the first obstacle obs1 is misdetected, the vehicle is not likely to suddenly stop based on the misdetected speed of the first obstacle obs1 whose collision distance is farther. However, the first obstacle obs2 whose speed is misdetected and whose collision distance is smaller is likely to cause the vehicle to suddenly stop.

[0085] According to the technical solution provided in the example, since the collision distance threshold is determined, the first obstacle whose speed is misdetected and which is not likely to cause the vehicle to suddenly stop can be filtered, thereby improving the accuracy of the target obstacle.

[0086] In addition, since the collision distance threshold can be determined based on the minimum collision distance of the second obstacle whose speed is not misdetected, the collision distance threshold is not a fixed value, but can be dynamically determined according to the actual situation of each obstacle around the vehicle at the target time, thereby further improving the accuracy of the collision distance threshold and the accuracy of the determined target obstacle.

[0087] In operation S440, the target obstacle causing the vehicle to have a predetermined behavior is determined from the at least one first obstacle according to the relevant position information of the first obstacle related to the target time and the distance threshold.

[0088] In an example, the relevant position information can include an estimated trajectory of the obstacle determined by an automatic driving algorithm of the vehicle at the target time, and the distance threshold can include a collision distance threshold. Accordingly, the first obstacle can be determined as the target obstacle in a case that the collision distance of the first obstacle is less than or equal to the collision distance threshold.

[0089] The present example can accurately obtain the target obstacle that has a greater probability of causing the vehicle to suddenly brake by screening the target obstacle through the distance threshold.

[0090] In another example, the relevant position information includes position point information (x obs , y obs ) of the obstacle at the target time, and the position information of the obstacle can be transformed from a world coordinate system to a vehicle body coordinate system to obtain transformed position point information (x obs , y obs ) of the obstacle, where the horizontal coordinate x obs represents the position of the obstacle in the vehicle width direction (i.e., the x direction), and the vertical coordinate y obs represents the position of the obstacle in the vehicle length direction (i.e., the y direction). Accordingly, the distance threshold can include a horizontal threshold x th and / or a vertical threshold y th .

[0091] For example, the obstacle can be determined as the target obstacle in a case that the absolute value of the horizontal coordinate x obs is less than or equal to the horizontal threshold x th , and / or the absolute value of the vertical coordinate y obs is less than or equal to the vertical threshold y th .

[0092] For another example, the obstacle can be determined as the target obstacle in a case that the absolute value of the horizontal coordinate x obs is less than or equal to the horizontal threshold x th , and / or the absolute value of the vertical coordinate y obs is less than or equal to a target value, where the target value is the smaller one of the vertical threshold y th and the collision distance threshold d th .

[0093] It can be understood that for the obstacle that is relatively close to the vehicle, the speed of the obstacle is prone to false detection, which is likely to cause the vehicle to suddenly brake. The present example can accurately determine the target obstacle by determining whether the obstacle is the target obstacle through the position point information (x obs , y obs ) of the obstacle.

[0094] Figure 5A - Figure 5B is a schematic diagram of a method for determining an obstacle according to an embodiment of the present disclosure.

[0095] The following combination Figure 5A - Figure 5B The method for determining obstacles that cause a predetermined behavior, as provided in this embodiment, will be described.

[0096] It can record in real time the abnormal moments and sensor reports of vehicles performing planned actions. For example, a time queue can be used to record vehicle sensor reports and the vehicle's planned trajectory in real time. If the vehicle takes measures such as sudden braking or lane changing, the current moment when the planned action occurs is recorded as the abnormal moment t. s The time delay between the vehicle sensing a hazard and executing a braking command is set to be less than t2 but greater than t1.

[0097] Next, determine the vehicle at (t) s -t2,t s Obstacles sensed within the time period -t1 are identified as candidate obstacles. Calculate (t s -t2,t s -t1) The straight-line distance between the location points of each candidate obstacle and the location point of the vehicle at each time point within the time period, and the straight-line distance less than or equal to the threshold L. s Obstacles located in front of the vehicle are added to a list of N obstacles and stored in the format of "time, obstacle ID, vehicle coordinate system position". The obstacle IDs are used for statistical analysis to obtain the N obstacles corresponding to each target time. The method for determining whether a candidate obstacle is located in front of the vehicle can be found above and will not be repeated here.

[0098] Next, for the N obstacles retained above, it is determined whether the obstacle's speed is a false alarm based on the time queue of the perceived reporting information. For example, at time t... s When +T, traverse N obstacles and determine whether each obstacle satisfies the following condition: within (t) s -t2,t s A non-zero velocity was detected during the time interval -t1), and in (t... s -T,t s -t2) time period and (t s , t s The velocity of the obstacle detected during the +T time period is 0, and the obstacle's velocity during the t time period is... s The position at time T and t s If the position change at time +T is less than the movement threshold, it can be determined that the obstacle's speed has been falsely reported; otherwise, it can be determined that the obstacle's speed has not been falsely reported. Through the above operation, the first obstacle 501 with false speed reports and the second obstacle 502 with no false speed reports can be obtained.

[0099] Next, for the first obstacle 501 and the second obstacle 502, it can be determined whether the obstacle and the vehicle exist collision risk, and the collision distance 503 is calculated in the case of existing collision risk, and the calculation method is as follows:

[0100] The planning trajectory of the vehicle at the reference time can be determined first The reference time can be the target time, or t s The time t2. For (t s t2, t s For a target time t in the period of (t The trajectory of the static obstacle is a point. Then, based on the planning trajectory of the vehicle and the estimated trajectory of the obstacle, it is determined whether the vehicle and the obstacle exist collision risk.

[0101] For example, the line segment connecting the estimated trajectory points of the obstacle can be traversed first, and if there is a line segment intersection, it is determined that the planning trajectory of the vehicle and the estimated trajectory of the obstacle intersect, and the intersection point is the collision risk position, for example Figure 5B The intersection point of the estimated trajectory of the obstacle 5012 and the planning trajectory of the vehicle is P2.

[0102] For example, if there is no trajectory line segment intersection, the intersection relationship between the two-dimensional bounding box and the planning trajectory of the vehicle can be further calculated, and the bounding box is obtained by expanding each point in the estimated trajectory of the obstacle.

[0103] If the edge line segment of the bounding box intersects the planning trajectory of the vehicle, or the distance from the vertex of the bounding box to the planning trajectory of the vehicle is less than the second distance threshold, it can be determined that the vehicle and the obstacle exist collision risk, and the sum of half of the vehicle width and the distance allowance can be taken as the second distance threshold. For example, the projection of the trajectory point of the obstacle to the trajectory line segment of the vehicle can be taken as the estimated collision position, for example, the intersection point of the estimated trajectory of the obstacle 5011 and the planning trajectory of the vehicle is P1.

[0104] If the distance from the vertex of the bounding box to the planning trajectory of the vehicle is greater than or equal to the second distance threshold, it is determined that the obstacle and the vehicle do not exist collision risk.

[0105] For example, the i-th obstacle at the target time t exists collision risk with the vehicle, and the collision distance can be determined according to the straight line distance between the collision risk position and the position of the vehicle 503, otherwise the collision distance 503 is recorded as infinity. For example Figure 5BAs shown, the collision distances 503 for obstacles 5011, 5012, 5013, and 5014 are OP1, OP2, OP3, and OP4, respectively. This can be recorded by setting the time t, obstacle ID, and collision distance. The collision distance of the obstacle at the recorded moment is saved in the format of "503".

[0106] Next, we can target (t) s -t2,t s For each target time t within the time period -t1), based on the collision distance 503 of the second obstacle 502 that did not cause a false speed detection at that target time t, a collision distance threshold 504 corresponding to that target time t is determined. This collision distance threshold 504 can be used to filter out the first obstacle 501 that does not cause sudden braking of the vehicle at the target time t. For example, for the target time t, the collision distance 503 of the second obstacle 502 that did not cause a false speed detection at that target time t can be used. The minimum value is determined as the collision distance threshold for that target at that moment. 504.

[0107] Next, the distance threshold corresponding to the target time t is used. The target obstacle 505 is determined from the N obstacles corresponding to the target time t. For example, based on the position of the obstacle in the vehicle coordinate system, different methods can be selected to determine whether the obstacle is the target obstacle 505 that will cause the vehicle to brake suddenly.

[0108] For example, at the target time t, the position (x, y) of the obstacle in the vehicle coordinate system satisfies L s <|x|<x 阈值 and If the obstacle exhibits a false velocity detection, then the obstacle can be identified as the target obstacle 505. In some embodiments, it is not required that the position point (x, y) satisfy L. s <|x|.

[0109] For example, at the target time t, the collision distance 503 of the obstacle is less than the collision distance threshold corresponding to that target time t. If 504 is selected, then the obstacle can be identified as the target obstacle. 505.

[0110] It can be seen that the embodiments of the present disclosure determine whether the obstacle causes the predetermined behavior of the vehicle such as sudden braking by combining the driving behavior of the vehicle and the detected obstacle information. In actual application, after the target obstacle 505 is determined by the above scheme, the cause of the obstacle speed false alarm can be analyzed, and the type of the obstacle can be labeled to obtain training samples. The automatic driving model is trained using the training samples, and then the automatic driving model is deployed on the vehicle side or the cloud side, thereby improving the accuracy of the automatic driving model in speed detection.

[0111] Figure 6 is a schematic structural block diagram of a device for determining an obstacle according to an embodiment of the present disclosure.

[0112] As shown in Figure 6 the device 600 for determining an obstacle causing a predetermined behavior can include a first determination module 610, a second determination module 620, a third determination module 630, and a fourth determination module 640.

[0113] The first determination module 610 is configured to determine a target time according to an abnormal time when the vehicle performs a predetermined behavior.

[0114] The second determination module 620 is configured to determine N obstacles from a plurality of candidate obstacles according to the target time, position information of the vehicle, and position information of each of the plurality of candidate obstacles; N is an integer greater than or equal to 1.

[0115] The third determination module 630 is configured to determine whether the speed of each of the N obstacles is false detected at the target time to obtain at least one first obstacle with speed false detection.

[0116] The fourth determination module 640 is configured to determine a target obstacle causing the vehicle to perform a predetermined behavior from the at least one first obstacle according to relevant position information of the first obstacle related to the target time and a distance threshold.

[0117] According to another embodiment of the present disclosure, the third determination module includes a first determination submodule configured to determine whether the speed of the obstacle is false detected at the target time according to at least one of the speed and position change information of the obstacle within a target period; wherein the start time of the target period is before the target time, and the end time of the target period is after the target time.

[0118] According to another embodiment of the present disclosure, the target period includes a first period, a second period, a third period, a first time and a second time. The first determining sub-module includes a determining unit configured to determine that the speed of the obstacle is misdetected in response to detecting that the obstacle is in a stationary state at the first period, in a non-stationary state at the second period, in a stationary state at the third period, and the distance between the position at the first time and the position at the second time is less than a moving threshold. The first period is before the target time. The second period is between the end time of the first period and the abnormal time, and the second period includes the target time. The third period is after the abnormal time, the first time is before the target time, and the second time is after the target time.

[0119] According to another embodiment of the present disclosure, the second determining module includes at least one of a first adding sub-module and a second adding sub-module. The first adding sub-module is configured to add a candidate obstacle in the plurality of candidate obstacles, which has a straight-line distance between the target time and the vehicle less than or equal to a first distance threshold, to the N obstacles. The second adding sub-module is configured to add a candidate obstacle in the plurality of candidate obstacles, which satisfies a predetermined relative position relationship with the vehicle at the target time, to the N obstacles.

[0120] According to another embodiment of the present disclosure, the relevant position information includes position point information of the first obstacle at the target time. The distance threshold includes a lateral threshold and a longitudinal threshold. The fourth determining module includes a second determining sub-module configured to determine the first obstacle as the target obstacle in a case where the distance between the position point and a first coordinate axis in the vehicle body coordinate system is less than the lateral threshold, and the distance between the position point and a second coordinate axis in the vehicle body coordinate system is less than the longitudinal threshold.

[0121] According to another embodiment of the present disclosure, the device further includes a fifth determining module and a sixth determining module. The fifth determining module is configured to determine whether the vehicle and the obstacle have a collision risk according to a planned trajectory of the vehicle at a reference time and an estimated trajectory of the obstacle at the target time. The sixth determining module is configured to determine a collision distance corresponding to the obstacle according to the estimated collision position and the position of the vehicle at the reference time in a case where it is determined that there is a collision risk.

[0122] According to another embodiment of the present disclosure, the relevant position information includes an estimated trajectory of the first obstacle at the target time, and the distance threshold includes a collision distance threshold. The fourth determining module includes a third determining sub-module configured to determine the first obstacle as the target obstacle in a case where the collision distance of the first obstacle is less than or equal to the collision distance threshold.

[0123] According to another embodiment of the present disclosure, the device further includes a seventh determining module configured to determine the collision distance threshold according to a minimum value of the collision distance of the second obstacle. The second obstacle is an obstacle in the N obstacles that does not have a speed misdetection.

[0124] According to another embodiment of the present disclosure, the fifth determining module is configured to determine that the vehicle and the obstacle are at risk of collision in a case where one of the following conditions is met: the planned trajectory and the estimated trajectory intersect; the planned trajectory and the estimated trajectory do not intersect, and the planned trajectory intersects with an edge of a bounding box, the bounding box being obtained by expanding a trajectory point in the estimated trajectory; and the planned trajectory and the estimated trajectory do not intersect, and a distance between the planned trajectory and a vertex of the bounding box is less than a second distance threshold.

[0125] In the technical solution of the present disclosure, the collection, storage, use, processing, transmission, provision and disclosure of user personal information involved in the technical solution comply with relevant laws and regulations and do not violate public order and good customs.

[0126] In the technical solution of the present disclosure, the authorization or consent of the user is obtained before the user personal information is acquired or collected.

[0127] According to an embodiment of the present disclosure, the present disclosure further provides an electronic device, comprising at least one processor; and a memory connected with the at least one processor in communication; the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to perform the method for determining the obstacle causing the predetermined behavior.

[0128] According to an embodiment of the present disclosure, the present disclosure further provides a non-transitory computer-readable storage medium storing computer instructions, wherein the computer instructions are used to enable a computer to perform the method for determining the obstacle causing the predetermined behavior.

[0129] According to an embodiment of the present disclosure, the present disclosure further provides a computer program product comprising a computer program, the computer program being executed by a processor to implement the method for determining the obstacle causing the predetermined behavior.

[0130] According to an embodiment of the present disclosure, the present disclosure further provides an autonomous vehicle comprising the electronic device. In actual application, the autonomous vehicle determines a target obstacle based on the predetermined behavior after generating a predetermined behavior such as emergency braking at a certain time, so as to optimize the autonomous driving model deployed by itself, so as to optimize the subsequent autonomous driving process, so as to reduce the number of times of the subsequent predetermined behavior such as emergency braking.

[0131] Figure 7A schematic block diagram of an example electronic device 700 that can be used to implement embodiments of the present disclosure is shown. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device may also represent various forms of mobile devices, such as personal digital processors, cellular phones, smartphones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely illustrative and are not intended to limit the implementation of the present disclosure described and / or claimed herein.

[0132] like Figure 7 As shown, device 700 includes a computing unit 701, which can perform various appropriate actions and processes based on a computer program stored in read-only memory (ROM) 702 or a computer program loaded from storage unit 708 into random access memory (RAM) 703. RAM 703 may also store various programs and data required for the operation of device 700. The computing unit 701, ROM 702, and RAM 703 are interconnected via bus 704. Input / output (I / O) interface 705 is also connected to bus 704.

[0133] Multiple components in device 700 are connected to I / O interface 705, including: input unit 706, such as keyboard, mouse, etc.; output unit 707, such as various types of monitors, speakers, etc.; storage unit 708, such as disk, optical disk, etc.; and communication unit 709, such as network card, modem, wireless transceiver, etc. Communication unit 709 allows device 700 to exchange information / data with other devices through computer networks such as the Internet and / or various telecommunications networks.

[0134] The computing unit 701 can be various general and / or special purpose processing components with processing and computing capabilities. Some examples of the computing unit 701 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various specialized artificial intelligence (AI) computing chips, various computing units running machine learning model algorithms, a digital signal processor (DSP), and any suitable processor, controller, microcontroller, etc. The computing unit 701 performs various methods and processes described above, such as the method of determining an obstacle causing a predetermined behavior. For example, in some embodiments, the method of determining an obstacle causing a predetermined behavior can be implemented as a computer software program tangibly embodied in a machine-readable medium, such as the storage unit 708. In some embodiments, part or all of the computer program can be loaded and / or installed onto the device 700 via the ROM 702 and / or the communication unit 709. When the computer program is loaded onto the RAM 703 and executed by the computing unit 701, one or more steps of the method of determining an obstacle causing a predetermined behavior described above can be performed. Alternatively, in other embodiments, the computing unit 701 can be configured to perform the method of determining an obstacle causing a predetermined behavior by any other suitable means, such as by means of firmware.

[0135] Various implementations of the systems and techniques described above can be realized in digital electronic circuitry, integrated circuitry, a field programmable gate array (FPGA), an application specific integrated circuit (ASIC), a system on a chip (SOC), a complex programmable logic device (CPLD), computer hardware, firmware, software, and / or combinations thereof. These various implementations can include implementation in one or more computer programs that are executable and / or interpretable on a programmable system including at least one programmable processor, which can be special or general purpose, coupled to receive data and instructions from, and to transmit data and instructions to, a storage system, at least one input device, and at least one output device.

[0136] Program code for carrying out methods of the present disclosure can be written in any combination of one or more programming languages. The program code can be provided to a processor or controller of a general purpose computer, special purpose computer, or other programmable data processing apparatus to produce a machine, such that the program code, when executed by the processor or controller, produces a means for implementing the functions / acts specified in the flowcharts and / or block diagrams. The program code can be executed entirely on a machine, partially on a machine, partially on a machine and partially on a remote machine or entirely on a remote machine or server.

[0137] In the context of this disclosure, a machine-readable medium can be a tangible medium that contains or stores a program for use by or in connection with an instruction execution system, apparatus, or device. The machine-readable medium can be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium can include but is not limited to an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any suitable combination of the foregoing. More specific examples of the machine-readable storage medium would include an electrical connection based on one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or Flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0138] To provide for interaction with a user, the systems and techniques described here can be implemented on a computer having a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user and a keyboard and a pointing device (e.g., a mouse or a trackball) by which the user can provide input to the computer. Other kinds of devices can be used to provide for interaction with a user as well; for example, feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form, including acoustic, speech, or tactile input.

[0139] The systems and techniques described here can be implemented in a computing system that includes a back end component (e.g., as a data server), or that includes a middleware component (e.g., an application server), or that includes a front end component (e.g., a user computer having a graphical user interface or a Web browser through which a user can interact with an implementation of the systems and techniques described here), or any combination of such back end, middleware, or front end components. The components of the system can be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include a local area network (LAN), a wide area network (WAN), and the Internet.

[0140] The computing system can include clients and servers. A client and server are generally remote from each other and typically interact through a communication network. The relationship of client and server arises by virtue of computer programs running on the respective computers and having a client-server relationship to each other.

[0141] It should be understood that the various forms of flow shown above can be used to reorder, add, or remove steps. For example, the steps described in the present disclosure can be performed in parallel, in series, or in a different order, as long as the desired results of the technology disclosed in the present disclosure are achieved, which is not limited herein.

[0142] The specific implementation described above does not constitute a limitation on the protection scope of the present disclosure. Those skilled in the art should understand that various modifications, combinations, sub-combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent replacements, and improvements made within the spirit and principles of the present disclosure shall be included in the protection scope of the present disclosure.

Claims

1. A method for determining an obstacle causing a predetermined behavior of a vehicle, comprising: determining a target time according to an abnormal time when the vehicle has the predetermined behavior; determining N obstacles from a plurality of candidate obstacles according to the target time, position information of the vehicle, and position information of each of the plurality of candidate obstacles; N is an integer greater than or equal to 1; determining whether a speed of each of the N obstacles is misdetected at the target time to obtain at least one first obstacle with speed misdetected; and determining a target obstacle causing the vehicle to have the predetermined behavior from the at least one first obstacle according to relevant position information of the at least one first obstacle related to the target time and a distance threshold. The determination of whether the speed of each of the N obstacles is misdetected at the target time comprises: determining whether the speed of the obstacle is misdetected at the target time according to at least one of speed and position change information of the obstacle within a target period; 2. The method of claim 1, wherein, wherein a start time of the target period is before the target time, and an end time of the target period is after the target time. The target period comprises a first period, a second period, a third period, a first time, and a second time; The determination of whether the speed of the obstacle is misdetected at the target time according to at least one of speed and position change information of the obstacle within a target period comprises:

3. The method of claim 2, wherein, in response to detecting that the obstacle is in a stationary state at the first period, in a non-stationary state at the second period, in a stationary state at the third period, and a distance between a position at the first time and a position at the second time is less than a movement threshold, determining that the speed of the obstacle is misdetected; wherein the first period is before the target time, the second period is between an end time of the first period and the abnormal time, the second period comprises the target time, the third period is after the abnormal time, the first time is before the target time, and the second time is after the target time. The determination of the N obstacles from the plurality of candidate obstacles according to the target time, the position information of the vehicle, and the position information of each of the plurality of candidate obstacles comprises at least one of:

4. The method of claim 1, wherein, adding a candidate obstacle in the plurality of candidate obstacles, which has a straight-line distance between the target time and the vehicle less than or equal to a first distance threshold, to the N obstacles; and adding a candidate obstacle in the plurality of candidate obstacles, which satisfies a predetermined relative position relationship with the vehicle at the target time, to the N obstacles. The relevant position information comprises position point information of the first obstacle at the target time; the distance threshold comprises a lateral threshold and a longitudinal threshold; and the determination of the target obstacle causing the vehicle to have the predetermined behavior from the at least one first obstacle according to the relevant position information of the at least one first obstacle related to the target time and the distance threshold comprises: ​ 5. The method of claim 1, wherein, ​ In a case where it is determined that the distance between the position point and a first coordinate axis in the vehicle body coordinate system is less than the lateral threshold value, and the distance between the position point and a second coordinate axis in the vehicle body coordinate system is less than the longitudinal threshold value, the first obstacle is determined as the target obstacle.

6. The method of any one of claims 1 to 5, further comprising: determining, according to a planned trajectory of the vehicle at a reference time and an estimated trajectory of the obstacle at the target time, whether there is a collision risk between the vehicle and the obstacle; and in a case where it is determined that there is a collision risk, determining a collision distance corresponding to the obstacle according to an estimated collision position and a position of the vehicle at the reference time.

7. The method of claim 6, wherein, The relevant position information includes an estimated trajectory of the first obstacle at the target time, and the distance threshold value includes a collision distance threshold value; and the determining, according to the relevant position information of the at least one first obstacle related to the target time and the distance threshold value, of the target obstacle causing the predetermined behavior of the vehicle from the at least one first obstacle includes: in a case where it is determined that the collision distance of the first obstacle is less than or equal to the collision distance threshold value, the first obstacle is determined as the target obstacle.

8. The method of claim 7, further comprising: determining the collision distance threshold value according to a minimum value of the collision distances of the second obstacles; wherein the second obstacles are obstacles in the N obstacles that do not have a speed false detection.

9. The method of claim 6, wherein, In a case where it is determined that one of the following conditions is met, it is determined that there is a collision risk between the vehicle and the obstacle: the planned trajectory and the estimated trajectory intersect; the planned trajectory and the estimated trajectory do not intersect, and the planned trajectory intersects with an edge of a bounding box, the bounding box being obtained by expanding a trajectory point in the estimated trajectory; and and the planned trajectory and the estimated trajectory do not intersect, and a distance between the planned trajectory and a vertex of the bounding box is less than a second distance threshold value.

10. An apparatus for determining an obstacle causing a predetermined behavior, comprising: a first determining module configured to determine a target time according to an abnormal time at which a vehicle has a predetermined behavior; a second determining module configured to determine N obstacles from a plurality of candidate obstacles according to the target time, position information of the vehicle, and respective position information of the plurality of candidate obstacles; N is an integer greater than or equal to 1; a third determining module configured to determine whether a speed of each of the N obstacles has a false detection at the target time to obtain at least one first obstacle having a speed false detection; and a fourth determining module configured to determine a target obstacle causing the predetermined behavior of the vehicle from the at least one first obstacle according to relevant position information of the at least one first obstacle related to the target time and a distance threshold value. The third determining module comprises:

11. The apparatus of claim 10, wherein, a first determining submodule configured to determine whether the speed of the obstacle has a false detection at the target time according to at least one of speed and position change information of the obstacle within a target time period. ​ The starting moment of the target time period is before the target moment, and the ending moment of the target time period is after the target moment.

12. The apparatus of claim 11, wherein, The target time period includes a first time period, a second time period, a third time period, a first moment, and a second moment; and the first determining sub-module includes: The determining unit is configured to, in response to detecting that the obstacle is in a static state during the first time period, in a non-static state during the second time period, in a static state during the third time period, and the distance between the position at the first moment and the position at the second moment is less than a moving threshold, determine that the speed of the obstacle is subject to a false detection. The first time period is before the target moment, the second time period is between the ending moment of the first time period and the abnormal moment, the second time period includes the target moment, the third time period is after the abnormal moment, the first moment is before the target moment, and the second moment is after the target moment.

13. The apparatus of claim 10, wherein, The second determining module includes at least one of: A first adding sub-module configured to add, from the plurality of candidate obstacles, a candidate obstacle whose straight-line distance between the target moment and the vehicle is less than or equal to a first distance threshold, to the N obstacles; and A second adding sub-module configured to add, from the plurality of candidate obstacles, a candidate obstacle that satisfies a predetermined relative position relationship with the vehicle at the target moment, to the N obstacles. The relevant position information includes position point information of the first obstacle at the target moment.

14. The apparatus of claim 10, wherein, The distance threshold includes a lateral threshold and a longitudinal threshold; and the fourth determining module includes: A second determining sub-module configured to, in a case where it is determined that the distance between the position point and a first coordinate axis in a vehicle body coordinate system is less than the lateral threshold, and the distance between the position point and a second coordinate axis in the vehicle body coordinate system is less than the longitudinal threshold, determine that the first obstacle is a target obstacle.

15. The apparatus according to any one of claims 10 to 14, further comprising: A fifth determining module configured to determine, according to a planned trajectory of the vehicle at a reference moment and an estimated trajectory of the obstacle at the target moment, whether the vehicle and the obstacle are at risk of collision; and A sixth determining module configured to, in a case where it is determined that there is a risk of collision, determine a collision distance corresponding to the obstacle according to an estimated collision position and a position of the vehicle at the reference moment. The relevant position information includes an estimated trajectory of the first obstacle at the target moment, and the distance threshold includes a collision distance threshold; and the fourth determining module includes: A third determining sub-module configured to, in a case where it is determined that the collision distance of the first obstacle is less than or equal to the collision distance threshold, determine that the first obstacle is a target obstacle.

16. The apparatus of claim 15, wherein, 17. The apparatus according to claim 16, further comprising: A seventh determining module configured to determine the collision distance threshold according to a minimum value of collision distances of second obstacles; The second obstacles are obstacles from the N obstacles that are not subject to a false detection of speed. ​ ​ 18. The apparatus of claim 15, wherein, The fifth determining module is configured to determine that the vehicle and the obstacle are at risk of collision in a case where it is determined that one of the following conditions is met: the planned trajectory and the estimated trajectory intersect; the planned trajectory and the estimated trajectory do not intersect, and the planned trajectory intersects with an edge of a bounding box, the bounding box being obtained by expanding a trajectory point in the estimated trajectory; and the planned trajectory and the estimated trajectory do not intersect, and a distance between the planned trajectory and a vertex of the bounding box is less than a second distance threshold.

19. An electronic device, comprising: at least one processor; and a memory connected to the at least one processor in communication; wherein the memory stores instructions executable by the at least one processor, the instructions being executed by the at least one processor to enable the at least one processor to perform the method of any one of claims 1 to 9. The computer instructions are configured to enable the computer to perform the method of any one of claims 1 to 9.

20. A non-transitory computer readable storage medium having stored thereon computer instructions, wherein, 21. An autonomous vehicle comprising the electronic device of claim 19. ​

Images