Intersection passing method, apparatus, device, and medium

Abstract

The application discloses a kind of intersection passing method, device, equipment and medium, comprising: when target vehicle reaches target position, the planned trajectory of each lane of the target vehicle from starting road to target road is determined, wherein the target position is the position with the stop line of the starting road There is a preset distance;Determine the initial expectation corresponding to each planning track;Obtain the predicted trajectory of the first vehicle driving influence object, and determine the first target planning track from each planning track with the predicted trajectory There is interactive influence;The first target planning track is updated, and the updated expectation corresponding to the first target planning track is obtained;Based on the current expectation of each planning track, determine the optimal trajectory from each planning track, and drive the target vehicle based on the optimal trajectory to pass through the current intersection.In this way, the efficiency of intersection passing can be improved.

Description

Technical Field

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

[0002] Currently, when a vehicle enters a target lane at an intersection, there are multiple lane options. The number of lane options corresponds to the number of lanes on the target road. Different lane options correspond to different driving efficiencies. Under the premise of safety, efficient and fast passage through intersections is a common demand of drivers.

[0003] Therefore, how to improve the efficiency of crossing an intersection is a technical problem that those skilled in the art need to solve. Summary of the Invention

[0004] In view of this, the purpose of this application is to provide a method, device, equipment and medium for passing through an intersection, which can improve the efficiency of passing through an intersection. The specific solution is as follows:

[0005] In a first aspect, the present application discloses a method for passing an intersection, comprising:

[0006] When the target vehicle reaches a target position, determining a planned trajectory of the target vehicle from the starting road to each lane of the target road, wherein the target position is a position at a preset distance from a stop line of the starting road;

[0007] Determine the initial expectation corresponding to each planned trajectory;

[0008] Obtaining a predicted trajectory of a first vehicle driving-affecting object, and determining a first target planned trajectory that interacts with the predicted trajectory from among the planned trajectories;

[0009] Performing an expectation update on the first target planning trajectory to obtain an updated expectation corresponding to the first target planning trajectory;

[0010] An optimal trajectory is determined from each planned trajectory based on current expectations of each planned trajectory, and the target vehicle is driven based on the optimal trajectory to pass through the current intersection.

[0011] Optionally, obtaining a predicted trajectory of the first vehicle driving-affecting object and determining a first target planned trajectory that interacts with the predicted trajectory from each planned trajectory includes:

[0012] Obtaining a first predicted trajectory of a first motor vehicle on the starting road based on an intersection passing manner of the target vehicle, wherein the intersection passing manner includes left turn, right turn, and straight ahead, and when the intersection passing manner is a left turn, the first motor vehicle is a motor vehicle on a left-turn lane; when the intersection passing manner is a right turn, the first motor vehicle is a motor vehicle on a right-turn lane; and when the intersection passing manner is straight ahead, the first motor vehicle is a motor vehicle on a straight ahead lane;

[0013] If the first predicted trajectory of the first motor vehicle does not extend to the target road, binding a target lane of the target road to the first motor vehicle based on the lane corresponding to the intersection passing mode on the starting road and each lane on the target road;

[0014] In a case where the first predicted trajectory of the first motor vehicle extends to the target road, the lane of the target road to which the first predicted trajectory extends is used as the target lane corresponding to the first motor vehicle;

[0015] Based on the lanes in the target road and the lanes in the starting road corresponding to each planned trajectory and the target lane in the target road and the lane in the starting road corresponding to the first motor vehicle, a first target planned trajectory that interacts with the first predicted trajectory is determined from each planned trajectory.

[0016] Optionally, obtaining a predicted trajectory of the first vehicle driving-affecting object and determining a first target planned trajectory that interacts with the predicted trajectory from each planned trajectory includes:

[0017] Obtaining a second predicted trajectory of a target object, wherein the target object includes a pedestrian, a non-motor vehicle, and a second motor vehicle;

[0018] Determining whether there is a possibility of collision between the target vehicle and the target object on the corresponding planned trajectory based on the second predicted trajectory and each planned trajectory;

[0019] A planned trajectory with a possible collision is determined as a first target planned trajectory that interacts with the second predicted trajectory.

[0020] Optionally, performing expected updating on the first target planning trajectory to obtain an updated expectation corresponding to the first target planning trajectory includes:

[0021] Determining a post-interaction speed planning curve corresponding to the first target planning trajectory based on the interaction influence;

[0022] The first target planning trajectory is updated based on the original speed curve corresponding to the first target planning trajectory and the post-interaction speed planning curve to obtain an updated expectation corresponding to the first target planning trajectory.

[0023] Optionally, also include:

[0024] In the case where it is determined that a vehicle behind the target vehicle has occupied the road, determining a second target planned trajectory from the planned trajectories;

[0025] performing an expectation update on the second target planning trajectory based on the number of lane changes on the target road corresponding to the second target planning trajectory relative to the desired lane, to obtain an updated expectation corresponding to the second target planning trajectory;

[0026] The desired lane is a lane determined based on the way the target road passes through the next intersection after passing the current intersection.

[0027] Optionally, determining an optimal trajectory from each planned trajectory based on current expectations of each planned trajectory, and driving the target vehicle based on the optimal trajectory to pass through the current intersection, includes:

[0028] Determining an optimal trajectory from each planned trajectory based on current expectations of each planned trajectory, and driving the target vehicle based on the optimal trajectory;

[0029] Obtaining a predicted trajectory of the second vehicle's driving influencing object to obtain a target predicted trajectory;

[0030] Determining whether there is a possibility of collision between the target vehicle and the second vehicle driving influencing object based on the target predicted trajectory and the optimal trajectory;

[0031] If there is a possibility of collision between the target vehicle and the second vehicle driving-affecting object, the optimal trajectory is corrected to obtain a corrected trajectory, wherein the corrected trajectory is consistent with a lane in the target road corresponding to the optimal trajectory;

[0032] Calculating an expectation corresponding to the corrected trajectory based on the corrected trajectory and the optimal trajectory;

[0033] A new optimal trajectory is determined from the corrected trajectory and other planned trajectories, and the target vehicle is continued to be driven based on the new optimal trajectory to pass the current intersection.

[0034] Optionally, also include:

[0035] If the corrected trajectory cannot be obtained, speed planning is performed on the current optimal trajectory to continue driving the target vehicle through the current intersection based on the current optimal trajectory and the planned speed.

[0036] In a second aspect, the present application discloses a road intersection passing device, comprising:

[0037] a planned trajectory determination module, configured to determine a planned trajectory of the target vehicle from a starting road to each lane of the target road when the target vehicle reaches a target position, wherein the target position is a position at a preset distance from a stop line of the starting road;

[0038] An initial expectation determination module is used to determine the initial expectation corresponding to each planned trajectory;

[0039] an impact trajectory determination module, configured to obtain a predicted trajectory of a first vehicle driving impact object, and determine a first target planned trajectory from each planned trajectory that has an interactive impact with the predicted trajectory;

[0040] an expectation updating module, configured to update the expectation of the first target planning trajectory to obtain an updated expectation corresponding to the first target planning trajectory;

[0041] The optimal trajectory determination module is used to determine the optimal trajectory from each planned trajectory based on the current expectations of each planned trajectory, and to drive the target vehicle based on the optimal trajectory to pass the current intersection.

[0042] In a third aspect, the present application discloses an electronic device, including a memory and a processor, wherein:

[0043] The memory is used to store computer programs;

[0044] The processor is used to execute the computer program to implement the aforementioned intersection passing method.

[0045] In a fourth aspect, the present application discloses a computer-readable storage medium for storing a computer program, wherein the computer program implements the aforementioned intersection passing method when executed by a processor.

[0046] In a fifth aspect, the present application provides a computer program product, comprising a computer program / instruction, which implements the steps of the aforementioned intersection passing method when executed by a processor.

[0047] It can be seen from the above scheme that the present application provides a method for passing through an intersection, including: when a target vehicle arrives at a target position, determining a planned trajectory of the target vehicle from a starting road to each lane in the target road, wherein the target position is a position at a preset distance from the stop line of the starting road; determining an initial expectation corresponding to each planned trajectory; obtaining a predicted trajectory of an object affected by the driving of a first vehicle, and determining a first target planned trajectory that has an interactive influence with the predicted trajectory from each planned trajectory; performing an expectation update on the first target planned trajectory to obtain an updated expectation corresponding to the first target planned trajectory; determining an optimal trajectory from each planned trajectory based on the current expectation of each planned trajectory, and driving the target vehicle based on the optimal trajectory to pass through the current intersection.

[0048] It can be seen that the beneficial effects of this application are as follows: when about to pass through an intersection, a planned trajectory is performed from the starting road to each lane in the target road, and the initial expectation corresponding to each planned trajectory is determined. The expectation of the corresponding planned trajectory is updated based on the driving influencing factors. The optimal trajectory is determined from each planned trajectory based on the expectation of each planned trajectory, and then the vehicle is driven along the optimal trajectory to pass through the current intersection. In this way, trajectory planning is first performed and the initial expectation of each planned trajectory is determined. Then, the expectation of the trajectory is adjusted according to the driving influencing factors. The trajectory with the highest efficiency in passing the intersection can be determined from each trajectory, and driving can be performed, which can improve the efficiency of passing the intersection.

[0049] Correspondingly, the intersection passing device, equipment and readable storage medium provided by this application also have the above-mentioned technical effects. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are merely embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on the provided drawings without any creative work.

[0051] Figure 1 A flow chart of a method for passing an intersection provided in an embodiment of the present application;

[0052] Figure 2 A schematic diagram of an initial trajectory planning provided in an embodiment of the present application;

[0053] Figure 3 A lane binding schematic diagram provided in an embodiment of the present application;

[0054] Figure 4 A lane binding schematic diagram provided in an embodiment of the present application;

[0055] Figure 5 A lane binding schematic diagram provided in an embodiment of the present application;

[0056] Figure 6 A lane binding schematic diagram provided in an embodiment of the present application;

[0057] Figure 7 A lane binding schematic diagram provided in an embodiment of the present application;

[0058] Figure 8 A lane binding schematic diagram provided in an embodiment of the present application;

[0059] Figure 9 A lane binding schematic diagram provided in an embodiment of the present application;

[0060] Figure 10 A lane binding schematic diagram provided in an embodiment of the present application;

[0061] Figure 11 A trajectory impact diagram provided in an embodiment of the present application;

[0062] Figure 12 A trajectory interaction schematic diagram is provided for an embodiment of the present application;

[0063] Figure 13 Another trajectory interaction schematic diagram provided in an embodiment of the present application;

[0064] Figure 14 A schematic diagram of trajectory correction provided in an embodiment of the present application;

[0065] Figure 15 A flow chart for passing an intersection provided in an embodiment of the present application;

[0066] Figure 16 A schematic diagram of the structure of an intersection passing device provided in an embodiment of the present application;

[0067] Figure 17 A structural diagram of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0068] The following will be combined with the drawings in the embodiments of this application to clearly and completely describe the technical solutions in the embodiments of this application. Obviously, the embodiments described are only part of the embodiments of this application, not all of the embodiments. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of this application.

[0069] See also Figure 1 As shown, the embodiment of the present application discloses a method for passing through an intersection, including:

[0070] Step S11: When the target vehicle reaches a target position, a planned trajectory of the target vehicle from the starting road to each lane in the target road is determined, wherein the target position is a position at a preset distance from a stop line of the starting road.

[0071] The target vehicle is the currently traveling vehicle, and the starting road is the road the target vehicle is currently on. The planned trajectory is the trajectory from the target vehicle's current lane on the starting road to each lane on the target road. The planned trajectory can be a virtual lane centerline, i.e., the line connecting the centerline of the target vehicle's current lane and the centerlines of each lane on the target road. When the target vehicle reaches the target location, a planned trajectory is generated from the starting road to each lane on the target road. For example, if the preset distance is 50 meters and the target vehicle is 50 meters from the stop line on the starting road, the planned trajectory of the target vehicle from the starting road to each lane on the target road is determined.

[0072] Step S12: Determine the initial expectation corresponding to each planned trajectory.

[0073] Among them, expectation can represent the passing efficiency corresponding to the trajectory. The higher the expectation, the higher the passing efficiency.

[0074] In an optional implementation, the initial expectation corresponding to each planned trajectory may be determined based on the planned trajectory length and the target road length.

[0075] Among them, the planned trajectory length and the length of the planned trajectory can be determined based on the starting point and the end point of the planned trajectory being the stop line of the starting road and the stop line of the target road on the current intersection side.

[0076] The target road length is the length of the target road, that is, the length of the road between the stop line of the current intersection and the stop line of the next intersection. The target road length and the lane of the target road selected when passing the current intersection affect the passing efficiency of the next intersection. For example, if the next intersection requires a right turn and the target road is shorter, selecting the leftmost lane of the target road when passing the current intersection will affect the passing efficiency of the next intersection. The embodiment of the present application takes the next intersection into consideration when passing the current intersection, thereby ensuring that the target vehicle has a high passing efficiency when passing multiple intersections continuously.

[0077] The embodiment of the present application can create a map at the intersection (taking left turn as an example, straight and right turns can be obtained in the same way). The map will include the target road pointed to by the starting road. Road passing method. For the ego vehicle, that is, the target vehicle, there is The method is as follows. n represents the number of left-turn lanes on the road where the vehicle is located, and m represents the number of lanes on the target road. The trajectory coordinate system can be established to establish the initial trajectory expectation. Figure 2 As shown, Figure 2 This is a schematic diagram of an initial trajectory planning example provided by this application. Before the vehicle enters L1-L4, a coordinate system is established 50 meters before the stop line, with the vehicle's head direction y perpendicular to the vehicle's head direction x. L1-L4 represent the planned trajectory. An initial expected value Ei is assigned to each Li based on the lengths of L1, L2, L3, and L4, as well as the target road length. i can be 1, 2, 3, 4, and so on.

[0078] ;

[0079] ;

[0080] Among them, ki is the proportional coefficient, You can use the preset value, such as 0.8, is the number of lane changes for each lane on the target road relative to the desired lane (for example, if the vehicle turns right after reaching the target road, the number of lane changes corresponding to L4 is 0, and the number of lane changes corresponding to L3 is 1. If the vehicle continues to go straight, the desired lane can be L2 or L3, that is, the desired lane is determined by the way the next intersection is passed in the next driving route), longi is the length of Li, is the target road length.

[0081] Step S13: obtaining a predicted trajectory of the first vehicle driving influencing object, and determining a first target planned trajectory that has an interactive impact with the predicted trajectory from each planned trajectory.

[0082] Among them, the first vehicle driving influencing object can be understood as an object that may affect the driving of the target vehicle, such as other motor vehicles, non-motor vehicles, pedestrians, etc.

[0083] In an optional embodiment, the first predicted trajectory of the first motor vehicle in the starting road can be obtained based on the intersection passing method of the target vehicle, wherein the intersection passing method includes left turn, right turn and straight going. When the intersection passing method is left turn, the first motor vehicle is a motor vehicle on the left turn lane; when the intersection passing method is right turn, the first motor vehicle is a motor vehicle on the right turn lane; when the intersection passing method is straight going, the first motor vehicle is a motor vehicle on the straight lane.

[0084] Among them, the lanes corresponding to the intersection passing methods are: if the intersection passing method is left turn, the corresponding lane is the left turn lane; if the intersection passing method is right turn, the corresponding lane is the right turn lane; if the intersection passing method is straight ahead, the corresponding lane is the straight ahead lane.

[0085] In a case where the first predicted trajectory of the first motor vehicle does not extend to the target road, a target lane in the target road is bound to the first motor vehicle based on the lane corresponding to the intersection passing mode in the starting road and each lane in the target road.

[0086] The first motor vehicle may be a motor vehicle included in the lane where the target vehicle is currently located and in front of the target vehicle, as well as a motor vehicle in a lane corresponding to other said intersection passing methods. If it is a motor vehicle in the lane where the target vehicle is currently located and behind the target vehicle, it may be ignored.

[0087] In an optional embodiment, the lane in the target road corresponding to the lane in the initial road can be determined based on the number of lanes corresponding to the intersection passing method, the number of lanes in the target road, and the lane arrangement order, so as to bind the target lane in the target road to the first motor vehicle.

[0088] Assume that n is the number of lanes corresponding to the intersection passing mode of the current road, and m is the number of lanes of the target road, and the maximum value can be 5. When m=n, the lanes of the target road can be mapped one-to-one with the lanes of the current road, and the lanes can be mapped in order from left lane to right lane. Figure 3 As shown, Figure 3 A lane binding diagram provided in an embodiment of the present application is provided. Figure 3 Take m=n=2 as an example, and the same applies to other values. When m>n, it can include the following cases: If n=1, each lane of the target road is the target lane corresponding to the lane of the current road, see Figure 4 As shown, Figure 4 A lane binding diagram provided in the embodiment of this application. If n=2, m=3, see Figure 5 As shown, Figure 5 A lane binding diagram provided in an embodiment of the present application. If n=2, m=4, and m is an integer multiple of n, the aforementioned integer multiple lanes can be allocated to the lanes corresponding to the intersection passing mode of each current road. Figure 6 As shown, Figure 6 A lane binding diagram provided in an embodiment of the present application. If n=2, m=5: See Figure 7 As shown, Figure 7 This is a lane binding diagram provided by the embodiment of this application. If n=3, m=4: See Figure 8 As shown, Figure 8 This is a lane binding diagram provided by the embodiment of this application. If n=3, m=5: See Figure 9 As shown, Figure 9 This is a lane binding diagram provided by the embodiment of this application. If n=4, m=5: See Figure 10As shown Figure 10 This is a schematic diagram of lane binding provided by an embodiment of the present application. If n > m, it is only necessary to search in reverse order according to the example corresponding to m > n.

[0089] When the first prediction trajectory of the first motor vehicle extends to the target road, the lane in the target road to which the first prediction trajectory extends is used as the target lane corresponding to the first motor vehicle.

[0090] It should be noted that the prediction trajectory can be a trajectory predicted for a preset future duration. As the object moves, it can successively be in a state of not extending to the target road and a state of extending to the target road.

[0091] Based on the lanes in the target roads corresponding to each planned trajectory, the lanes in the starting road, and the target lane in the target road corresponding to the first motor vehicle, the lanes in the starting road, a first target planned trajectory that has an interactive influence with the first prediction trajectory is determined from each planned trajectory.

[0092] In an embodiment of the present application, when m1 < n1 and m2 < n2 or m1 > n1 and m2 > n2 are satisfied, it is determined that there is no interactive influence without considering the interactive influence, otherwise there is an interactive influence, where m1 represents the lane of any planned trajectory in the starting road, m2 represents the target lane of the any planned trajectory in the target road, n1 represents the lane of the first prediction trajectory in the starting road, and n2 represents the target lane of the first prediction trajectory in the target road. The lanes in the starting road and the target road can be numbered in order from the left lane to the right lane, and m1, m2, n1, and n2 can be the corresponding lane numbers obtained by numbering. For example, the lane numbers in the starting road can be 1, 2, 3, 4, 5, etc. from left to right, and the lane numbers in the target road can be 1, 2, 3, 4, 5, etc. from left to right.

[0093] In an optional implementation manner, a second prediction trajectory of a target object can be obtained, where the target object includes pedestrians, non-motor vehicles, and a second motor vehicle; based on the second prediction trajectory and each planned trajectory, it is determined whether there is a possible collision between the target vehicle and the target object on the corresponding planned trajectory; the planned trajectory with a possible collision is determined as the first target planned trajectory that has an interactive influence with the second prediction trajectory.

[0094] The second motor vehicle can be a motor vehicle other than the first motor vehicle that may have an impact on the target vehicle. For example, a motor vehicle that is currently turning left at the intersection and whose trajectory on the road opposite the initial road extends to the intersection is considered to be a potential collision. If a first preset collision condition is met, a collision is determined to be possible; otherwise, no collision is determined to be possible. The first preset collision condition can be an intersection between the second predicted trajectory and the planned trajectory, or a distance between two points on the two trajectories at the same time that is less than a preset threshold. These two conditions can be selected based on vehicle size, etc.

[0095] In this way, by classifying the objects at the intersection that may affect the driving of the target vehicle and handling them according to the situation, the traffic situation at the intersection can be reflected more realistically, thereby determining the optimal trajectory more accurately.

[0096] Step S14: performing an expectation update on the first target planning trajectory to obtain an updated expectation corresponding to the first target planning trajectory.

[0097] In an optional embodiment, the post-interaction speed planning curve corresponding to the first target planning trajectory can be determined based on the interaction influence; the first target planning trajectory is updated based on the original speed curve corresponding to the first target planning trajectory and the post-interaction speed planning curve to obtain the updated expectation corresponding to the first target planning trajectory.

[0098] The embodiment of the present application can determine the post-interaction speed planning curve corresponding to the first target planning trajectory after determining the first target planning trajectory that has an interactive influence with the predicted trajectory. For example, it can decelerate or accelerate after deceleration. Among them, the post-interaction speed planning curve and the original speed curve are both speed value change curves of the target vehicle in the future, and the vehicle travels according to the speed value described in the speed curve. By integrating the difference between the original speed curve and the post-interaction speed planning curve, the impact value corresponding to the interaction is obtained, and the expectation is updated based on the impact value. In this way, the expectation of the planned trajectory with interactive influence can be reduced, thereby ensuring the accurate selection of the optimal trajectory.

[0099] That is, the embodiment of the present application can update the trajectory expectation based on the predicted trajectory of objects such as vehicles. Taking left turn as an example, the expectation can be updated based on the predicted trajectory of the vehicle in the left turn lane of the road and whether the trajectory of non-stationary obstacles invades the planned trajectory. When the predicted trajectory of the vehicle in the left turn lane does not extend to the target road, it is bound to the target lane corresponding to its left turn lane, see Figure 11 As shown, Figure 11A trajectory influence diagram is provided for an embodiment of the present application. The target lane corresponding to the purple vehicle is the lane pointed by the red arrow. L4 and L3 interact with the trajectory of the purple vehicle, so the interaction influence is calculated. The interaction influence refers to mapping the future trajectory of the purple vehicle and the planned trajectory of the vehicle onto the ST diagram. When the purple vehicle affects the speed of the vehicle, the expectation will be recalculated in the trajectory selection. The target lane, i.e., the optimal trajectory, is reselected based on the current expectation. The ST diagram plots the distance (s) that the obstacle moves in the next time (t) in a two-dimensional coordinate system, and the vehicle trajectory can be obtained. The expectation can be expressed by the following expression:

[0100] ;

[0101] ;

[0102] in, is the original speed curve, is the speed planning curve after interaction.

[0103] If the obs (obstacle) vehicle's predicted trajectory extends to the target lane when turning left, the purple vehicle's target lane is replaced with its extended lane, and the road expectation is recalculated on the ST map (consistent with the above expectation expression).

[0104] In addition, for target objects including pedestrians, non-motor vehicles, and second motor vehicles, such as pedestrians running red lights, the ST map is updated according to their predicted trajectories. The road expectation is re-updated.

[0105] ;

[0106] ;

[0107] It is the velocity curve planned when the trajectory corresponding to the target interacts with the vehicle. This formula can be combined with the previous one.

[0108] In an optional embodiment, when it is determined that there is a vehicle behind the target vehicle occupying the road, a second target planning trajectory is determined from each planned trajectory; based on the number of lane changes on the target road corresponding to the second target planning trajectory relative to the expected lane, the second target planning trajectory is updated to obtain an updated expectation corresponding to the second target planning trajectory; wherein, the expected lane is a lane determined based on the intersection passing method at the next intersection after the target road passes the current intersection.

[0109] If the preset rear vehicle occupying road conditions are met, it is determined that a rear vehicle behind the target vehicle is occupying the road. If the intersection crossing method is a left turn, the preset rear vehicle occupying road conditions are: the target vehicle is in the leftmost left-turn lane and there is a vehicle behind it, where "behind" refers to the target vehicle's lane, behind it, or the vehicle is not in the leftmost left-turn lane, but there are other vehicles in the leftmost left-turn lane. If the intersection crossing method is a right turn, the preset rear vehicle occupying road conditions are: the target vehicle is in the rightmost right-turn lane and there is a vehicle behind it, where "behind" refers to the target vehicle's lane, behind it, or the vehicle is not in the rightmost right-turn lane, but there are other vehicles in the rightmost right-turn lane.

[0110] For example, after the target road passes the current intersection, the intersection passing method at the next intersection is to turn right, and the expected lane is the rightmost lane. The intersection passing method at the next intersection is to turn left, and the expected lane is the leftmost lane. The intersection passing method at the next intersection is to go straight, and the expected lane is the straight road.

[0111] That is, the embodiment of the present application can consider the situation where the vehicle behind the vehicle in the lane seizes the road and update the expectation. The logic corresponding to the vehicle behind the vehicle seizing the road can be added to the expectation function. Because the driver may lower the set speed when driving an autonomous vehicle, this means that other vehicles are faster than the vehicle behind the driver, that is, the vehicle behind the driver may overtake the vehicle at the intersection. Once the vehicle behind the driver overtakes the vehicle behind the driver, that is, the vehicle behind the driver occupies the road in front of the driver, the driver will lose part of the expected road. Therefore, when calculating the expectation, it is necessary to determine the right of way and update the expectation. The formula used is as follows:

[0112] ;

[0113] in, It indicates the number of lane changes of each lane on the target road relative to the desired lane. It can be understood that each lane on the target road corresponds to a different planning trajectory. For example, we can first calculate ,when When it is not equal to 1, K4 is set to 0 (because if there are multiple lane changes from the leftmost lane, the desired lane is the leftmost lane, and the leftmost lane right has no effect on the current choice). =1, if the preset road-occupying condition for the rear vehicle is met, K4 takes the preset value; if the preset road-occupying condition for the rear vehicle is not met, K4 takes 0. vo represents the current vehicle speed, and 45 represents 45 km / h, meaning that the higher the speed, the smaller the impact of K4.

[0114] In this way, the situation of the rear vehicle occupying the road is taken into consideration, and the influencing factors can be considered more comprehensively, expectations can be updated, and the accuracy of the optimal lane can be guaranteed.

[0115] Step S15: determining an optimal trajectory from each planned trajectory based on the current expectations of each planned trajectory, and driving the target vehicle based on the optimal trajectory to pass the current intersection.

[0116] In an optional embodiment, an optimal trajectory can be determined from each planned trajectory based on the current expectations of each planned trajectory, and the target vehicle can be driven based on the optimal trajectory; the predicted trajectory of the second vehicle driving influencing object is obtained to obtain a target predicted trajectory; based on the target predicted trajectory and the optimal trajectory, it is judged whether there is a possibility of collision between the target vehicle and the second vehicle driving influencing object; if there is a possibility of collision between the target vehicle and the second vehicle driving influencing object, the optimal trajectory is corrected to obtain a corrected trajectory, wherein the corrected trajectory is consistent with the lane in the target road corresponding to the optimal trajectory; the expectation corresponding to the corrected trajectory is calculated based on the corrected trajectory and the optimal trajectory; a new optimal trajectory is determined from the corrected trajectory and other planned trajectories, and the target vehicle continues to drive based on the new optimal trajectory to pass the current intersection.

[0117] Among them, the optimal trajectory is the trajectory with the greatest current expectation. The second vehicle driving influencing object can be understood as an object that may have an impact on the driving of the target vehicle after the target vehicle crosses the stop line. It can be determined that there is a possibility of collision when the second preset collision condition is met, otherwise it can be determined that there is no possibility of collision, wherein the second preset collision condition can be that there is an intersection between the second predicted trajectory and the planned trajectory, or that the distance between two points on the two trajectories at the same time is less than a preset threshold. The above two conditions can be selected according to the size of the vehicle, etc. The corrected trajectory is a trajectory modified based on the optimal trajectory to meet the requirement of vehicle non-collision, and the starting lane and the end lane of the corrected trajectory are consistent with those of the optimal trajectory before correction.

[0118] In the embodiment of the present application, the expectation corresponding to the planned trajectory can be adjusted based on the corrected trajectory and the planned trajectory corresponding to the corrected trajectory as the expectation corresponding to the corrected trajectory.

[0119] In an alternative implementation, the expected update formula is as follows:

[0120] ;

[0121] ;

[0122] in, is the proportionality coefficient, represents the original trajectory, Indicates the correction trajectory, which can be a lateral position offset. represents the original speed curve, Indicates the replanned speed curve. represents the expected impact value of the corrected trajectory.

[0123] If the corrected trajectory cannot be obtained, speed planning is performed on the current optimal trajectory to continue driving the target vehicle through the current intersection based on the current optimal trajectory and the planned speed.

[0124] When there are multiple vehicles at an intersection, they may all affect the target vehicle's travel, and it may be impossible to obtain a corrected trajectory. In this case, speed planning can be performed on the current trajectory to ensure that the vehicle passes through the intersection safely.

[0125] See also Figure 12 As shown, the embodiment of the present application provides a trajectory interaction schematic diagram. Figure 13 As shown, Figure 13 Another trajectory interaction diagram provided for an embodiment of the present application. The ego vehicle (i.e., the target vehicle) is traveling at an intersection: it travels based on the optimal trajectory before the stop line, and when encountering an interaction point with the obs estimate at the intersection, it is expected to be updated again. The interaction point is determined based on the predicted trajectory (it can be a trajectory intersection point or a vehicle neighboring point), and is projected onto the ST diagram based on the distance between the two to the interaction point. The trajectory influencing factors are set and the trajectory of the ego vehicle is corrected. Usually, the ego vehicle plans the left speed on the ST diagram. Each vehicle will give a predicted trajectory of 6s, which is split by time 0.1 to obtain 60 trajectory points. The ego vehicle can also plan a trajectory of 6s, and 60 points can be obtained by the same method. The scenario of the two vehicles meeting in the future can be simulated based on time t. According to calculations, the two vehicles will collide after 6s (which will not actually happen). Therefore, the t of the two vehicles for the interaction point (traversal time t, when the distance between the two vehicles is less than dmin for the first time) is known, and the vehicle trajectory is offset by x centimeters based on the interaction point. See Figure 14 As shown, Figure 14 A schematic diagram of trajectory correction provided for an embodiment of the present application. The red dot is the point corresponding to the interaction point time point t on the original trajectory. After moving the point to the yellow dot (the yellow dot and the red dot are x centimeters apart, in a direction perpendicular to the original trajectory), the new trajectory obtained is the corrected trajectory. The expectation corresponding to the corrected trajectory is updated, and the optimal trajectory is reselected. When there are multiple vehicles at the intersection, if the corrected trajectory corresponding to the yellow dot cannot be obtained (for example, the left and right vehicles are traveling together towards the vehicle path (i.e., the current driving trajectory)), the corrected trajectory calculated according to the above method will jump up and down, and the road priority will not be determined, but speed planning will be performed on the existing path. Further, travel according to the selected trajectory.

[0126] See also Figure 15 As shown, Figure 15A flowchart for passing an intersection is provided for an embodiment of the present application. The initial expectation of each planned trajectory can be obtained first, and then it is determined whether there is an interaction effect, and then whether there is a trajectory intersection, and whether there is a rear vehicle occupying the road, and the trajectory expectation is updated. After the stop line, the optimal trajectory is traversed based on the selected road, that is, obs. If there is an interaction point, the trajectory is offset to obtain a corrected trajectory, the expectation is updated, and the optimal trajectory is reselected. Among them, the vehicle refers to the first motor vehicle, and the non-vehicle refers to the target object, and the target object includes non-vehicles. Figure 15 Take non-vehicles as an example.

[0127] As can be seen, when the embodiment of the present application is about to pass through an intersection, it plans trajectories for each lane from the starting road to the target road, determines the initial expectation corresponding to each planned trajectory, updates the expectation of the corresponding planned trajectory based on the driving influencing factors, determines the optimal trajectory from each planned trajectory based on the expectation of each planned trajectory, and then drives along the optimal trajectory to pass the current intersection. In this way, by first planning the trajectory and determining the initial expectation of each planned trajectory, and then adjusting the trajectory expectation based on the driving influencing factors, it is possible to determine the trajectory with the highest intersection passing efficiency from each trajectory and drive on it, which can improve the efficiency of passing the intersection.

[0128] See also Figure 16 As shown, the embodiment of the present application discloses a road intersection passing device, comprising:

[0129] a planned trajectory determination module 11, configured to determine a planned trajectory of the target vehicle from a starting road to each lane of the target road when the target vehicle reaches a target position, wherein the target position is a position at a preset distance from a stop line of the starting road;

[0130] An initial expectation determination module 12 is used to determine the initial expectation corresponding to each planned trajectory;

[0131] An impact trajectory determination module 13 is configured to obtain a predicted trajectory of a first vehicle driving impact object and determine a first target planned trajectory that has an interactive impact with the predicted trajectory from among the planned trajectories;

[0132] An expectation updating module 14 is configured to update the expectation of the first target planning trajectory to obtain an updated expectation corresponding to the first target planning trajectory;

[0133] The optimal trajectory determination module 15 is configured to determine an optimal trajectory from the planned trajectories based on current expectations of the planned trajectories, and drive the target vehicle based on the optimal trajectory to pass through the current intersection.

[0134] In an optional implementation manner, the impact trajectory determination module 13 may be specifically configured to:

[0135] Obtaining a first predicted trajectory of a first motor vehicle on the starting road based on an intersection passing manner of the target vehicle, wherein the intersection passing manner includes left turn, right turn, and straight ahead, and when the intersection passing manner is a left turn, the first motor vehicle is a motor vehicle on a left-turn lane; when the intersection passing manner is a right turn, the first motor vehicle is a motor vehicle on a right-turn lane; and when the intersection passing manner is straight ahead, the first motor vehicle is a motor vehicle on a straight ahead lane;

[0136] If the first predicted trajectory of the first motor vehicle does not extend to the target road, binding a target lane of the target road to the first motor vehicle based on the lane corresponding to the intersection passing mode on the starting road and each lane on the target road;

[0137] In a case where the first predicted trajectory of the first motor vehicle extends to the target road, the lane of the target road to which the first predicted trajectory extends is used as the target lane corresponding to the first motor vehicle;

[0138] Based on the lanes in the target road and the lanes in the starting road corresponding to each planned trajectory and the target lane in the target road and the lane in the starting road corresponding to the first motor vehicle, a first target planned trajectory that interacts with the first predicted trajectory is determined from each planned trajectory.

[0139] In an optional implementation manner, the impact trajectory determination module 13 may be specifically configured to:

[0140] Obtaining a second predicted trajectory of a target object, wherein the target object includes a pedestrian, a non-motor vehicle, and a second motor vehicle;

[0141] Determining whether there is a possibility of collision between the target vehicle and the target object on the corresponding planned trajectory based on the second predicted trajectory and each planned trajectory;

[0142] A planned trajectory with a possible collision is determined as a first target planned trajectory that interacts with the second predicted trajectory.

[0143] Correspondingly, the expectation update module 14 can be specifically used to determine the post-interaction speed planning curve corresponding to the first target planning trajectory based on the interaction influence; based on the original speed curve corresponding to the first target planning trajectory and the post-interaction speed planning curve, the first target planning trajectory is expected to be updated to obtain the updated expectation corresponding to the first target planning trajectory.

[0144] In an optional embodiment, the impact trajectory determination module 13 can also be used to determine a second target planned trajectory from each planned trajectory when it is determined that there is a vehicle behind the target vehicle occupying the road; accordingly, the expectation update module 14 can also be used to perform an expectation update on the second target planned trajectory based on the number of lane changes on the target road corresponding to the second target planned trajectory relative to the expected lane, to obtain an updated expectation corresponding to the second target planned trajectory; wherein the expected lane is a lane determined based on the intersection passing method at the next intersection after the target road passes the current intersection.

[0145] In an optional embodiment, the optimal trajectory determination module 15 may be specifically configured to:

[0146] Determining an optimal trajectory from each planned trajectory based on current expectations of each planned trajectory, and driving the target vehicle based on the optimal trajectory;

[0147] Obtaining a predicted trajectory of the second vehicle's driving influencing object to obtain a target predicted trajectory;

[0148] Determining whether there is a possibility of collision between the target vehicle and the second vehicle driving influencing object based on the target predicted trajectory and the optimal trajectory;

[0149] If there is a possibility of collision between the target vehicle and the second vehicle driving-affecting object, the optimal trajectory is corrected to obtain a corrected trajectory, wherein the corrected trajectory is consistent with a lane in the target road corresponding to the optimal trajectory;

[0150] Calculating an expectation corresponding to the corrected trajectory based on the corrected trajectory and the optimal trajectory;

[0151] A new optimal trajectory is determined from the corrected trajectory and other planned trajectories, and the target vehicle is continued to be driven based on the new optimal trajectory to pass the current intersection.

[0152] If the corrected trajectory cannot be obtained, speed planning is performed on the current optimal trajectory to continue driving the target vehicle through the current intersection based on the current optimal trajectory and the planned speed.

[0153] As can be seen, when the embodiment of the present application is about to pass through an intersection, it plans trajectories for each lane from the starting road to the target road, determines the initial expectation corresponding to each planned trajectory, updates the expectation of the corresponding planned trajectory based on the driving influencing factors, determines the optimal trajectory from each planned trajectory based on the expectation of each planned trajectory, and then drives along the optimal trajectory to pass the current intersection. In this way, by first planning the trajectory and determining the initial expectation of each planned trajectory, and then adjusting the trajectory expectation based on the driving influencing factors, it is possible to determine the trajectory with the highest intersection passing efficiency from each trajectory and drive on it, which can improve the efficiency of passing the intersection.

[0154] See also Figure 17 As shown, an embodiment of the present application discloses an electronic device 20, including a processor 21 and a memory 22; wherein the memory 22 is used to store a computer program; the processor 21 is used to execute the computer program, the intersection passing method disclosed in the above embodiment.

[0155] For the specific process of the above-mentioned intersection passing method, please refer to the corresponding content disclosed in the above-mentioned embodiment, which will not be repeated here.

[0156] Furthermore, the memory 22 as a carrier for resource storage may be a read-only memory, a random access memory, a magnetic disk or an optical disk, etc., and the storage method may be temporary storage or permanent storage.

[0157] In addition, the electronic device 20 also includes a power supply 23, a communication interface 24, an input / output interface 25 and a communication bus 26; wherein the power supply 23 is used to provide an operating voltage for each hardware device on the electronic device 20; the communication interface 24 can create a data transmission channel between the electronic device 20 and an external device, and the communication protocol it follows is any communication protocol that can be applied to the technical solution of the present application, and is not specifically limited here; the input / output interface 25 is used to obtain external input data or output data to the outside world, and its specific interface type can be selected according to specific application needs and is not specifically limited here.

[0158] Furthermore, an embodiment of the present application also discloses a computer-readable storage medium for storing a computer program, wherein the computer program, when executed by a processor, implements the intersection passing method disclosed in the aforementioned embodiment.

[0159] For the specific process of the above-mentioned intersection passing method, please refer to the corresponding content disclosed in the above-mentioned embodiment, which will not be repeated here.

[0160] Furthermore, an embodiment of the present application also provides a computer program product, including a computer program / instruction, which, when executed by a processor, implements the steps of the intersection passing method disclosed in the aforementioned embodiment.

[0161] For the specific process of the above-mentioned intersection passing method, please refer to the corresponding content disclosed in the above-mentioned embodiment, which will not be repeated here.

[0162] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on its differences from the other embodiments. Reference can be made to the descriptions of the identical or similar parts between the various embodiments. For the devices disclosed in the embodiments, since they correspond to the methods disclosed in the embodiments, the descriptions are relatively simple, and the relevant parts can be referred to the descriptions of the methods.

[0163] The steps of the methods or algorithms described in conjunction with the embodiments disclosed herein may be implemented directly using hardware, a software module executed by a processor, or a combination of the two. The software module may be placed in random access memory (RAM), internal memory, read-only memory (ROM), electrically programmable ROM, electrically erasable programmable ROM, registers, a hard disk, a removable disk, a CD-ROM, or any other form of storage medium known in the art.

[0164] The above is a detailed introduction to the intersection passing method, device, equipment and medium provided by the present application. Specific examples are used in this article to illustrate the principles and implementation methods of the present application. The description of the above embodiments is only used to help understand the method of the present application and its core idea; at the same time, for general technical personnel in this field, based on the ideas of the present application, there will be changes in the specific implementation methods and application scope. In summary, the content of this specification should not be understood as a limitation on the present application.

Claims

1. A method for passing an intersection, characterized in that: include: When the target vehicle reaches a target position, determining a planned trajectory of the target vehicle from the starting road to each lane of the target road, wherein the target position is a position at a preset distance from a stop line of the starting road; Determine an initial expectation for each planned trajectory; wherein the expectation represents the passing efficiency corresponding to the trajectory, and the initial expectation for each planned trajectory is determined based on the planned trajectory length and the target road length. The planned trajectory length is the length of the planned trajectory between the stop line of the starting road and the stop line of the target road at the current intersection. The target road length is the length of the road from the stop line of the current intersection to the stop line of the next intersection. Obtaining a predicted trajectory of a first vehicle driving-affecting object, and determining a first target planned trajectory that interacts with the predicted trajectory from among the planned trajectories; Performing an expectation update on the first target planning trajectory to obtain an updated expectation corresponding to the first target planning trajectory; An optimal trajectory is determined from each planned trajectory based on current expectations of each planned trajectory, and the target vehicle is driven based on the optimal trajectory to pass through the current intersection.

2. The method for passing through an intersection according to claim 1, wherein: The obtaining of the predicted trajectory of the first vehicle driving influencing object and determining, from each planned trajectory, a first target planned trajectory that has an interactive impact on the predicted trajectory includes: Obtaining a first predicted trajectory of a first motor vehicle on the starting road based on an intersection passing manner of the target vehicle, wherein the intersection passing manner includes left turn, right turn, and straight ahead, and when the intersection passing manner is a left turn, the first motor vehicle is a motor vehicle on a left-turn lane; when the intersection passing manner is a right turn, the first motor vehicle is a motor vehicle on a right-turn lane; and when the intersection passing manner is straight ahead, the first motor vehicle is a motor vehicle on a straight ahead lane; If the first predicted trajectory of the first motor vehicle does not extend to the target road, binding a target lane of the target road to the first motor vehicle based on the lane corresponding to the intersection passing mode on the starting road and each lane on the target road; In a case where the first predicted trajectory of the first motor vehicle extends to the target road, the lane of the target road to which the first predicted trajectory extends is used as the target lane corresponding to the first motor vehicle; Based on the lanes in the target road and the lanes in the starting road corresponding to each planned trajectory and the target lane in the target road and the lane in the starting road corresponding to the first motor vehicle, a first target planned trajectory that interacts with the first predicted trajectory is determined from each planned trajectory.

3. The method for passing through an intersection according to claim 1, wherein: The obtaining of the predicted trajectory of the first vehicle driving influencing object and determining, from each planned trajectory, a first target planned trajectory that has an interactive impact on the predicted trajectory includes: Obtaining a second predicted trajectory of a target object, wherein the target object includes a pedestrian, a non-motor vehicle, and a second motor vehicle; Determining whether there is a possibility of collision between the target vehicle and the target object on the corresponding planned trajectory based on the second predicted trajectory and each planned trajectory; A planned trajectory with a possible collision is determined as a first target planned trajectory that interacts with the second predicted trajectory.

4. The method for passing through an intersection according to claim 1, wherein: The updating of the expectation of the first target planning trajectory to obtain an updated expectation corresponding to the first target planning trajectory includes: Determining a post-interaction speed planning curve corresponding to the first target planning trajectory based on the interaction influence; The first target planning trajectory is updated based on the original speed curve corresponding to the first target planning trajectory and the post-interaction speed planning curve to obtain an updated expectation corresponding to the first target planning trajectory.

5. The method for passing through an intersection according to claim 1, characterized in that: Also includes: In the case where it is determined that a vehicle behind the target vehicle has occupied the road, determining a second target planned trajectory from the planned trajectories; performing an expectation update on the second target planning trajectory based on the number of lane changes on the target road corresponding to the second target planning trajectory relative to the desired lane, to obtain an updated expectation corresponding to the second target planning trajectory; The desired lane is a lane determined based on the way the target road passes through the next intersection after passing the current intersection.

6. The method for passing through an intersection according to any one of claims 1 to 5, characterized in that: Determining an optimal trajectory from each planned trajectory based on current expectations of each planned trajectory, and driving the target vehicle based on the optimal trajectory to pass through the current intersection, including: Determining an optimal trajectory from each planned trajectory based on current expectations of each planned trajectory, and driving the target vehicle based on the optimal trajectory; Obtaining a predicted trajectory of the second vehicle's driving influencing object to obtain a target predicted trajectory; Determining whether there is a possibility of collision between the target vehicle and the second vehicle driving influencing object based on the target predicted trajectory and the optimal trajectory; If there is a possibility of collision between the target vehicle and the second vehicle driving-affecting object, the optimal trajectory is corrected to obtain a corrected trajectory, wherein the corrected trajectory is consistent with a lane in the target road corresponding to the optimal trajectory; Calculating an expectation corresponding to the corrected trajectory based on the corrected trajectory and the optimal trajectory; A new optimal trajectory is determined from the corrected trajectory and other planned trajectories, and the target vehicle is continued to be driven based on the new optimal trajectory to pass the current intersection.

7. The method for passing through an intersection according to claim 6, characterized in that: Also includes: If the corrected trajectory cannot be obtained, speed planning is performed on the current optimal trajectory to continue driving the target vehicle through the current intersection based on the current optimal trajectory and the planned speed.

8. A road intersection passing device, characterized in that: include: a planned trajectory determination module, configured to determine a planned trajectory of the target vehicle from a starting road to each lane of the target road when the target vehicle reaches a target position, wherein the target position is a position at a preset distance from a stop line of the starting road; An initial expectation determination module is configured to determine an initial expectation corresponding to each planned trajectory; wherein the expectation represents the passing efficiency corresponding to the trajectory, and the initial expectation corresponding to each planned trajectory is determined based on the planned trajectory length and the target road length. The planned trajectory length is the length of the planned trajectory between the stop line of the starting road and the stop line of the target road at the current intersection. The target road length is the length of the road from the stop line of the current intersection to the stop line of the next intersection. an impact trajectory determination module, configured to obtain a predicted trajectory of a first vehicle driving impact object, and determine a first target planned trajectory from each planned trajectory that has an interactive impact with the predicted trajectory; an expectation updating module, configured to update the expectation of the first target planning trajectory to obtain an updated expectation corresponding to the first target planning trajectory; The optimal trajectory determination module is used to determine the optimal trajectory from each planned trajectory based on the current expectations of each planned trajectory, and to drive the target vehicle based on the optimal trajectory to pass the current intersection.

9. An electronic device, characterized in that: comprising a memory and a processor, wherein: The memory is used to store computer programs; The processor is configured to execute the computer program to implement the intersection passing method according to any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that Used to store a computer program, wherein when the computer program is executed by a processor, the intersection passing method according to any one of claims 1 to 7 is implemented.

Images