An intersection lane selection method, device, equipment and readable storage medium

Abstract

Embodiments of the present application disclose a kind of intersection lane selection method, device, equipment and readable storage medium.The method comprises: obtaining the map data corresponding to current intersection, and navigation information, ego vehicle state information and obstacle state information;Based on the map data, the navigation information, the ego vehicle state information and the obstacle state information, determine the lane evaluation value of each candidate lane corresponding to the ego vehicle driving out the intersection, and for each candidate lane, determine the traffic evaluation value of the candidate lane based on the lane evaluation value of the candidate lane;Based on current trajectory planning algorithm and control tracking algorithm, determine the regulation and control cost of each candidate lane;According to the traffic evaluation value and regulation and control cost of each candidate lane, determine the optimal lane in each candidate lane.Application of the scheme provided in the embodiments of the present application can improve the accuracy of intersection lane selection of autonomous vehicle, and then ensure the driving safety of autonomous vehicle.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of automatic driving, in particular to a method and device for selecting a lane at an intersection, an apparatus and a readable storage medium. BACKGROUND

[0002] During the driving of an automatic driving vehicle, when the vehicle passes through an intersection (such as a crossroad, a roundabout, etc.), the exit lane that can be selected is usually multiple when the vehicle enters the intersection lane. In this case, since different road conditions are different, such as the size of traffic flow, the distribution of obstacles, etc., therefore, the selection of which lane to exit will directly affect the traffic efficiency, intersection safety and driving experience of the automatic driving vehicle.

[0003] The known method for selecting a lane at an intersection is usually based on a network to learn the optimal lane selection result from expert driving data. However, when learning the optimal lane selection result from expert driving data, the result is only based on the evaluation of the complete driving behavior of the driver in a specific scenario, and this result is not necessarily optimal for a specific automatic driving software system. Specifically, after selecting an exit lane, trajectory planning and control tracking need to be performed based on the selected lane, and trajectory planning and control tracking are completed by a separate trajectory generation module and a control tracking module. However, the trajectory generated by the trajectory generation module cannot guarantee to approach the expert trajectory, and the control tracking module cannot maintain a high control accuracy like an expert driver. This may cause the following problems: for a given optimal lane selection decision, there may be a situation where the vehicle collides with obstacles such as road edges, fences, barrels, etc. during the intersection due to unreasonable trajectory planning results or large control tracking errors. Therefore, how to improve the accuracy of the lane selection at the intersection of the automatic driving vehicle and thus ensure the driving safety of the automatic driving vehicle has become a technical problem to be solved. SUMMARY

[0004] The present application provides a method and device for selecting a lane at an intersection, an apparatus and a readable storage medium to improve the accuracy of the lane selection at the intersection of the automatic driving vehicle and thus ensure the driving safety of the automatic driving vehicle. The specific technical solutions are as follows.

[0005] In a first aspect, the present application provides a method for selecting a lane at an intersection, the method comprising:

[0006] obtaining map data corresponding to a current intersection, and navigation information, vehicle state information and obstacle state information;

[0007] determine a lane evaluation value of each candidate lane corresponding to the lane that the ego vehicle exits the intersection based on the map data, the navigation information, the ego vehicle state information, and the obstacle state information, and determine a passing evaluation value of each candidate lane based on the lane evaluation value of the candidate lane;

[0008] determine a regulation and control cost of each candidate lane based on the current trajectory planning algorithm and the control tracking algorithm;

[0009] determine an optimal lane from the candidate lanes based on the passing evaluation value and the regulation and control cost of each candidate lane.

[0010] In the embodiments of the present application, when selecting the lane at the intersection, not only the lane evaluation value of each candidate lane, that is, the current road condition of each candidate lane, is considered, but also the regulation and control cost of each candidate lane is considered based on the current trajectory planning algorithm and the control tracking algorithm. That is, if each candidate lane is used as the lane to exit the intersection, the corresponding planning and control result is considered, and the optimal lane is selected based on the lane evaluation value and the regulation and control cost of each candidate lane, which can ensure that the finally determined lane to exit the intersection is in an optimal condition and can meet the current planning and control algorithm power, thereby avoiding the collision between the autonomous vehicle and the obstacles such as the edge of the road, the fence, and the cone during the autonomous vehicle passing through the intersection, and improving the accuracy of the lane selection at the intersection of the autonomous vehicle, thereby ensuring the driving safety of the autonomous vehicle.

[0011] Optionally, before the step of determining the passing evaluation value of each candidate lane based on the lane evaluation value of the candidate lane, the method further comprises:

[0012] calculating a safety evaluation value of each candidate lane; the safety evaluation value includes a trajectory safety evaluation result and / or an intersection collision risk;

[0013] The step of determining the passing evaluation value of each candidate lane based on the lane evaluation value of the candidate lane comprises:

[0014] For each candidate lane, the passing evaluation value of the candidate lane is determined based on the lane evaluation value and the safety evaluation value of the candidate lane.

[0015] Optionally, when the safety evaluation value includes the trajectory safety evaluation result and the intersection collision risk, the step of calculating the safety evaluation value of each candidate lane comprises:

[0016] For each candidate lane, a planning trajectory from the current lane where the ego vehicle is located to the candidate lane is generated, and a trajectory safety evaluation result corresponding to the planning trajectory is calculated.

[0017] acquire a pre-stored intersection collision risk value corresponding to each intersection, and determine the intersection collision risk value corresponding to the current intersection according to the intersection collision risk value corresponding to each intersection.

[0018] Optionally, the step of determining the lane evaluation value of each candidate lane corresponding to the intersection that the ego vehicle exits according to the map data, the navigation information, the ego vehicle state information and the obstacle state information comprises:

[0019] determining a current lane in which the ego vehicle is located, and determining each candidate lane corresponding to the intersection that the ego vehicle exits and a target lane corresponding to the ego vehicle before reaching a next intersection according to the map data and the navigation information;

[0020] for each candidate lane, calculating a first lane evaluation value from the current lane to the candidate lane according to the ego vehicle state information and the obstacle state information, and calculating a second lane evaluation value from the candidate lane to the target lane according to the ego vehicle state information and the obstacle state information; the first lane evaluation value comprises a first traffic efficiency evaluation value and / or a first traffic safety evaluation value; and the second lane evaluation value comprises a second traffic efficiency evaluation value and / or a second traffic safety evaluation value;

[0021] for each candidate lane, calculating a lane evaluation value of the candidate lane according to the first lane evaluation value and the second lane evaluation value.

[0022] Optionally, the obstacle state information comprises static obstacle state information and dynamic obstacle state information; when the first lane evaluation value comprises a first traffic efficiency evaluation value and a first traffic safety evaluation value, the step of calculating, for each candidate lane, a first lane evaluation value from the current lane to the candidate lane according to the ego vehicle state information and the obstacle state information comprises:

[0023] for each candidate lane, determining a smooth continuous curve length from the current lane to the candidate lane, and determining a first traffic efficiency evaluation value from the current lane to the candidate lane based on the smooth continuous curve length; the first traffic efficiency evaluation value is negatively correlated with the smooth continuous curve length;

[0024] For each of the candidate lanes, a static structure risk coefficient corresponding to the candidate lane is calculated according to the static obstacle state information, a dynamic traffic risk coefficient corresponding to the candidate lane is calculated according to the ego vehicle state information and the dynamic obstacle state information, and a first traffic safety evaluation value of the current lane to the candidate lane is calculated based on the static structure risk coefficient and the dynamic traffic risk coefficient; the first traffic safety evaluation value is negatively correlated with the static structure risk coefficient and the dynamic traffic risk coefficient; the static structure risk coefficient is related to the relative position of the static obstacle and the lane and the probability of the static obstacle being observed; the dynamic traffic risk coefficient is related to the collision time of the ego vehicle and the dynamic obstacle;

[0025] A first lane evaluation value of the current lane to the candidate lane is calculated according to the first traffic efficiency evaluation value and the first traffic safety evaluation value; the first lane evaluation value is positively correlated with the first traffic efficiency evaluation value and the first traffic safety evaluation value.

[0026] Optionally, when the second lane evaluation value includes a second traffic efficiency evaluation value and a second traffic safety evaluation value, the step of calculating the second lane evaluation value of the candidate lane to the target lane according to the ego vehicle state information and the obstacle state information includes:

[0027] The minimum number of lane changes from the candidate lane to the target lane is determined, and the expected travel time and the expected lane change time of each lane passed by the ego vehicle when the ego vehicle passes from the candidate lane to the target lane through the minimum number of lane changes are calculated according to the ego vehicle state information and the obstacle state information; the expected travel time of any lane is related to the expected driving distance of the ego vehicle in the lane and the obstacle distribution of the lane; the expected lane change time of any lane is positively correlated with the lane change interval length of the lane;

[0028] A second traffic efficiency evaluation value of the candidate lane to the target lane is calculated according to the minimum number of lane changes, the expected travel time and the expected lane change time of each lane; the second traffic efficiency evaluation value is negatively correlated with the minimum number of lane changes, the expected travel time and the expected lane change time;

[0029] A second traffic safety evaluation value of the candidate lane to the target lane is calculated according to the minimum number of lane changes; the second traffic safety evaluation value is negatively correlated with the minimum number of lane changes;

[0030] A second lane evaluation value of the candidate lane to the target lane is calculated according to the second traffic efficiency evaluation value and the second traffic safety evaluation value; the second lane evaluation value is positively correlated with the second traffic efficiency evaluation value and the second traffic safety evaluation value.

[0031] Optionally, the step of calculating, for each of the candidate lanes, a lane evaluation value of the candidate lane according to the first lane evaluation value and the second lane evaluation value comprises:

[0032] determining, for each of the candidate lanes, a first weight of the first lane evaluation value and a second weight of the second lane evaluation value;

[0033] weighting the first lane evaluation value and the second lane evaluation value based on the first weight and the second weight, and taking the weighting result as the lane evaluation value of the candidate lane.

[0034] Optionally, the step of determining, based on the map data, the navigation information, the ego vehicle state information and the obstacle state information, the lane evaluation values of each candidate lane corresponding to the ego vehicle driving out of the intersection comprises:

[0035] inputting the map data, the navigation information, the ego vehicle state information and the obstacle state information into a lane selection model pre-trained to obtain probability values of each candidate lane corresponding to the ego vehicle driving out of the intersection; the lane selection model is pre-trained according to expert driving data;

[0036] taking the probability values of each candidate lane as the lane evaluation values of the corresponding candidate lanes.

[0037] Optionally, the step of determining, based on the current trajectory planning algorithm and the control tracking algorithm, the regulation and control cost of each candidate lane comprises:

[0038] counting a first historical lateral error of the trajectory planning algorithm and a second historical lateral error of the control tracking algorithm;

[0039] calculating the regulation and control cost of each candidate lane according to the first historical lateral error and the second historical lateral error, the regulation and control cost being positively correlated with the first historical lateral error and the second historical lateral error.

[0040] Optionally, the step of determining, based on the current trajectory planning algorithm and the control tracking algorithm, the regulation and control cost of each candidate lane comprises:

[0041] generating, for each of the candidate lanes, a plurality of candidate trajectories from the current lane of the ego vehicle to the candidate lane based on the trajectory planning algorithm;

[0042] For each of the candidate trajectories, a planning cost of the candidate trajectory is calculated, and the control tracking algorithm is used to control track the candidate trajectory to obtain a corresponding control error, and a tracking cost of the candidate trajectory is calculated according to the control error; the tracking cost is positively correlated with the control error;

[0043] For each of the candidate trajectories, a planning cost of the candidate trajectory is calculated, and the control tracking algorithm is used to control track the candidate trajectory to obtain a corresponding control error, and a tracking cost of the candidate trajectory is calculated according to the control error; the tracking cost is positively correlated with the control error;

[0044] The candidate trajectory with the lowest trajectory cost is determined as the optimal lane of the candidate lane.

[0045] Optionally, the step of determining the optimal lane in each of the candidate lanes according to the passing evaluation value and the regulation and control cost of each of the candidate lanes comprises:

[0046] In each of the candidate lanes, the candidate lane with a regulation and control cost meeting a preset requirement is determined as a candidate lane.

[0047] In each of the candidate lanes, the candidate lane with the highest passing evaluation value is determined as the optimal lane.

[0048] In a second aspect, an embodiment of the present application provides a lane selection device at an intersection, and the device comprises:

[0049] A data acquisition module is configured to acquire map data corresponding to a current intersection, and navigation information, self-vehicle state information and obstacle state information;

[0050] A passing evaluation value determination module is configured to determine lane evaluation values of each candidate lane through which the self-vehicle exits the intersection based on the map data, the navigation information, the self-vehicle state information and the obstacle state information, and determine a passing evaluation value of each of the candidate lanes based on the lane evaluation value of the candidate lane.

[0051] A regulation and control cost determination module is configured to determine regulation and control costs of each of the candidate lanes based on a current trajectory planning algorithm and a control tracking algorithm.

[0052] An optimal lane determination module is configured to determine an optimal lane in each of the candidate lanes according to the passing evaluation value and the regulation and control cost of each of the candidate lanes.

[0053] Optionally, the device further comprises:

[0054] A safety evaluation value calculation module is configured to calculate a safety evaluation value of each of the candidate lanes; the safety evaluation value comprises a trajectory safety evaluation result and / or an intersection collision risk.

[0055] The passing evaluation value determination module is specifically configured to:

[0056] For each candidate lane, a passing evaluation value of the candidate lane is determined based on a lane evaluation value and a safety evaluation value of the candidate lane.

[0057] Optionally, when the safety evaluation value includes a trajectory safety evaluation result and an intersection collision risk, the safety evaluation value calculation module is specifically configured to:

[0058] For each candidate lane, a planned trajectory from a current lane where the ego vehicle is located to the candidate lane is generated, and a trajectory safety evaluation result corresponding to the planned trajectory is calculated.

[0059] An intersection collision risk value corresponding to each intersection is acquired in advance, and the intersection collision risk corresponding to the current intersection is determined according to the intersection collision risk values corresponding to the intersections.

[0060] Optionally, the passing evaluation value determination module includes:

[0061] A lane determination sub-module is configured to determine a current lane where the ego vehicle is located, and determine each candidate lane corresponding to the intersection where the ego vehicle exits and a target lane corresponding to a next intersection before the ego vehicle reaches the next intersection according to the map data and the navigation information.

[0062] A lane evaluation value calculation sub-module is configured to, for each candidate lane, calculate a first lane evaluation value from the current lane to the candidate lane according to the ego vehicle state information and the obstacle state information, and calculate a second lane evaluation value from the candidate lane to the target lane according to the ego vehicle state information and the obstacle state information; the first lane evaluation value includes a first passing efficiency evaluation value and / or a first passing safety evaluation value; and the second lane evaluation value includes a second passing efficiency evaluation value and / or a second passing safety evaluation value.

[0063] A lane evaluation value combination sub-module is configured to, for each candidate lane, calculate a lane evaluation value of the candidate lane according to the first lane evaluation value and the second lane evaluation value.

[0064] Optionally, the obstacle state information includes static obstacle state information and dynamic obstacle state information; and when the first lane evaluation value includes a first passing efficiency evaluation value and a first passing safety evaluation value, the lane evaluation value calculation sub-module is specifically configured to:

[0065] For each of the candidate lanes, a smooth continuous curve length from the current lane to the candidate lane is determined, and a first traffic efficiency evaluation value from the current lane to the candidate lane is determined based on the smooth continuous curve length; the first traffic efficiency evaluation value is negatively correlated with the smooth continuous curve length;

[0066] For each of the candidate lanes, a static structure risk coefficient corresponding to the candidate lane is calculated according to the static obstacle state information, a dynamic traffic risk coefficient corresponding to the candidate lane is calculated according to the ego vehicle state information and the dynamic obstacle state information; and a first traffic safety evaluation value from the current lane to the candidate lane is calculated based on the static structure risk coefficient and the dynamic traffic risk coefficient; wherein the first traffic safety evaluation value is negatively correlated with the static structure risk coefficient and the dynamic traffic risk coefficient; the static structure risk coefficient is related to the relative position of the static obstacle and the lane and the probability of the static obstacle being observed; the dynamic traffic risk coefficient is related to the collision time of the ego vehicle and the dynamic obstacle;

[0067] A first lane evaluation value from the current lane to the candidate lane is calculated according to the first traffic efficiency evaluation value and the first traffic safety evaluation value; the first lane evaluation value is positively correlated with the first traffic efficiency evaluation value and the first traffic safety evaluation value.

[0068] Optionally, when the second lane evaluation value includes a second traffic efficiency evaluation value and a second traffic safety evaluation value, the lane evaluation value calculation submodule is specifically configured to:

[0069] A minimum number of lane changes from the candidate lane to the target lane is determined, and an expected travel time and an expected lane changing time corresponding to each lane through which the ego vehicle passes when passing from the candidate lane to the target lane through the minimum number of lane changes are calculated according to the ego vehicle state information and the obstacle state information; wherein the expected travel time corresponding to any lane is related to the expected driving distance of the ego vehicle in the lane and the obstacle distribution of the lane; the expected lane changing time corresponding to any lane is positively correlated with the lane changing interval length of the lane;

[0070] A second traffic efficiency evaluation value from the candidate lane to the target lane is calculated according to the minimum number of lane changes, the expected travel time and the expected lane changing time corresponding to each lane; the second traffic efficiency evaluation value is negatively correlated with the minimum number of lane changes, the expected travel time and the expected lane changing time;

[0071] A second traffic safety evaluation value from the candidate lane to the target lane is calculated according to the minimum number of lane changes; the second traffic safety evaluation value is negatively correlated with the minimum number of lane changes;

[0072] According to the second traffic efficiency evaluation value and the second traffic safety evaluation value, a second lane evaluation value of the candidate lane to the target lane is calculated; the second lane evaluation value is positively correlated with the second traffic efficiency evaluation value and the second traffic safety evaluation value.

[0073] Optionally, the lane evaluation value combination sub-module is specifically configured to:

[0074] For each candidate lane, a first weight of the first lane evaluation value and a second weight of the second lane evaluation value are determined;

[0075] Based on the first weight and the second weight, the first lane evaluation value and the second lane evaluation value are weighted, and the weighted result is taken as the lane evaluation value of the candidate lane.

[0076] Optionally, the traffic evaluation value determination module is specifically configured to:

[0077] The map data, the navigation information, the ego vehicle state information and the obstacle state information are input into a lane selection model trained in advance to obtain a probability value of each candidate lane corresponding to the ego vehicle driving out of the intersection; the lane selection model is obtained according to expert driving data in advance;

[0078] The probability value of each candidate lane is taken as the lane evaluation value of the corresponding candidate lane.

[0079] Optionally, the regulation and control cost determination module is specifically configured to:

[0080] The first historical lateral error of the trajectory planning algorithm and the second historical lateral error of the control tracking algorithm are counted;

[0081] According to the first historical lateral error and the second historical lateral error, the regulation and control cost of each candidate lane is calculated, and the regulation and control cost is positively correlated with the first historical lateral error and the second historical lateral error.

[0082] Optionally, the regulation and control cost determination module is specifically configured to:

[0083] For each candidate lane, a plurality of candidate trajectories from the current lane where the ego vehicle is located to the candidate lane are generated based on the trajectory planning algorithm;

[0084] For each candidate trajectory, a planning cost of the candidate trajectory is calculated, and the control tracking algorithm is used to control and track the candidate trajectory to obtain a corresponding control error, and a tracking cost of the candidate trajectory is calculated according to the control error; the tracking cost is positively correlated with the control error;

[0085] For each of the candidate trajectories, a trajectory cost of the candidate trajectory is calculated according to a planning cost and a tracking cost of the candidate trajectory; the trajectory cost is positively correlated with the planning cost and the tracking cost;

[0086] A candidate trajectory with the lowest trajectory cost is determined, and the trajectory cost of the candidate trajectory is taken as a regulation cost of the candidate lane.

[0087] Optionally, the optimal lane determination module is specifically configured to:

[0088] In each of the candidate lanes, a candidate lane with a regulation cost meeting a preset requirement is determined as a candidate lane for selection;

[0089] In each of the candidate lanes for selection, a candidate lane with the highest passing evaluation value is determined as the optimal lane.

[0090] In a third aspect, an embodiment of the present application provides a computer device, comprising a memory and a processor, and the memory and the processor are coupled;

[0091] The memory is configured to store one or more computer instructions;

[0092] The processor is configured to execute the one or more computer instructions to implement the intersection lane selection method according to the first aspect.

[0093] In a fourth aspect, an embodiment of the present application provides a computer readable storage medium, which stores one or more computer instructions, and the instructions are executed by a processor to implement the intersection lane selection method according to the first aspect.

[0094] In a fifth aspect, the present application provides a computer program product, which comprises a computer program, and the computer program is executed by a processor to implement the intersection lane selection method according to the first aspect. BRIEF DESCRIPTION OF DRAWINGS

[0095] In order to more clearly illustrate the technical solutions of the embodiments of the present application or the prior art, the following will briefly introduce the drawings needed to be used in the embodiments or the prior art description. Obviously, the drawings in the following description are only some embodiments of the present application. For those skilled in the art, other drawings can also be obtained without creative labor on the basis of these drawings.

[0096] Figure 1 A schematic diagram of intersection lane selection and trajectory planning and tracking results;

[0097] Figure 2 A schematic diagram of a flow of an intersection lane selection method in an embodiment of the present application;

[0098] Figure 3 A flowchart for determining a lane evaluation value of each candidate lane in an embodiment of the present application is shown in FIG. 2;

[0099] Figure 4 A schematic diagram of a target lane determination result in an embodiment of the present application is shown in FIG. 3;

[0100] Figure 5 A schematic diagram of another target lane determination result in an embodiment of the present application is shown in FIG. 4;

[0101] Figure 6 A schematic diagram of a minimum number of lane changes of each lane in an embodiment of the present application is shown in FIG. 5;

[0102] Figure 7 A schematic diagram of a structure of an intersection lane selection device provided in an embodiment of the present application is shown in FIG. 6;

[0103] Figure 8 A schematic diagram of a structure of a computer device provided in an embodiment of the present application is shown in FIG. 7. DETAILED DESCRIPTION

[0104] The technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only some of the embodiments of the present application, but not all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor fall within the scope of protection of the present application.

[0105] It should be noted that the terms “include” and “have” and any variations thereof in the embodiments of the present application and the drawings are intended to cover non-exclusive inclusion. For example, a process, method, system, product or device including a series of steps or units is not limited to the listed steps or units, but can optionally further include steps or units not listed, or can optionally further include other steps or units inherent to the process, method, product or device.

[0106] Known intersection lane selection methods are usually based on a network to learn optimal lane selection results from expert driving data. However, after selecting an exit lane, trajectory planning and control tracking need to be performed based on the selected lane. The trajectory planning and control tracking are completed by a separate trajectory generation module and a control tracking module. However, the trajectory generated by the trajectory generation module cannot guarantee to approach an expert trajectory, and the control tracking module cannot maintain a high control accuracy like an expert driver. This may cause the following problems: for a given optimal lane selection decision, there may be a situation where the vehicle collides with obstacles such as road edges, fences, and cone barrels during the intersection due to unreasonable trajectory planning results or excessive control tracking errors.

[0107] As shown in Figure 1 According to the navigation information, the autonomous vehicle 110 needs to go straight through the current intersection, and the lane where the autonomous vehicle is currently located, i.e., the entry lane, is lane A. According to the map data or real-time sensing data, it can be seen that the lanes available for selection at the exit of the intersection include lane A', lane B' and lane C'. The expert lane learned by the network is lane A', and the corresponding expert trajectory is trajectory 120. However, after determining that the exit lane is lane A', the planning trajectory obtained by the current trajectory planning algorithm of the autonomous vehicle is trajectory 130, and the tracking trajectory obtained after control tracking is trajectory 140. It can be seen that the final trajectory 140 has a very small distance from the static obstacle on the road edge, and when the autonomous vehicle travels according to the trajectory 140, a collision between the autonomous vehicle and the obstacle will occur.

[0108] It can be seen that relying only on the expert trajectory to select the exit lane cannot guarantee the safety of the autonomous vehicle. The computing power of the trajectory planning algorithm and the control tracking algorithm also needs to be considered. Based on this, the embodiments of the present application disclose a method, device and equipment for selecting an intersection lane and a readable storage medium to improve the accuracy of the intersection lane selection of the autonomous vehicle and thus ensure the driving safety of the autonomous vehicle. The embodiments of the present application will be described in detail below.

[0109] Figure 2 A flowchart of a method for selecting an intersection lane provided by the embodiments of the present application is shown in the figure. The method can be applied to an autonomous vehicle, and specifically, can be applied to a processor in the autonomous vehicle. As shown in the figure, the method can include the following steps: Figure 2

[0110] S210: Obtain map data corresponding to the current intersection, as well as navigation information, vehicle state information and obstacle state information.

[0111] During the driving of the autonomous vehicle, path planning and the like can be performed based on pre-stored or real-time loaded map data. The map data can include navigation maps, high-precision maps and the like, which are all possible. The above navigation information can be a driving route determined by the autonomous vehicle according to a start point and an end point set by a user; the vehicle state information and the obstacle state information can be obtained through sensing. Specifically, the vehicle state information can include, for example, position, speed, heading angle, acceleration and the like; the obstacles can include all dynamic obstacles (vehicles, pedestrians, etc.) and static obstacles (plants, road edges, cones, etc.) in the surrounding environment of the vehicle that can affect the driving of the vehicle, and each obstacle state information can include, for example, its position, speed and the like.

[0112] ​S220: determine lane evaluation values of each candidate lane corresponding to the exit of the intersection based on the map data, the navigation information, the ego vehicle state information and the obstacle state information, and determine a passing evaluation value of each candidate lane based on the lane evaluation value of the candidate lane.

[0113] When the autonomous vehicle drives to any intersection, according to the navigation information, it can need to drive in different directions after passing through the intersection, such as straight, left turn, right turn, etc. It can be understood that when the autonomous vehicle drives in any direction, the lanes that can be driven in (which can be referred to as candidate lanes) are usually multiple, and the conditions of each candidate lane are usually not exactly the same, such as the current traffic volume of each candidate lane is different, the distance from the current lane of the autonomous vehicle is different, etc., which will result in that after the autonomous vehicle selects different candidate lanes, the subsequent driving conditions (passing time, safety, etc.) are also different. In the embodiments of the present application, in order to improve the driving experience and safety of the user driving the autonomous vehicle, the optimal candidate lane corresponding to the exit of the intersection can be selected when the autonomous vehicle drives to the intersection.

[0114] Specifically, the autonomous vehicle can first determine the driving direction corresponding to the exit of the intersection according to the map data and the navigation information, and then each lane of the road in the direction is regarded as a candidate lane. Further, the lane evaluation values of each candidate lane can be determined based on the map data, the ego vehicle state information and the obstacle state information. That is, each candidate lane can be evaluated according to the road conditions of each candidate lane, and then the optimal candidate lane can be selected as the lane to be driven into according to the evaluation results of each candidate lane. For each candidate lane, the lane evaluation value can reflect the degree to which the candidate lane is suitable for the autonomous vehicle to pass through, and the higher the lane evaluation value, the more suitable the lane is for the autonomous vehicle to pass through.

[0115] In one implementation manner, as shown in FIG. 2, the process of determining the lane evaluation values of each candidate lane can include the following steps: Figure 3

[0116] S310: determine the current lane where the autonomous vehicle is located, and determine each candidate lane corresponding to the exit of the intersection and the target lane corresponding to the next intersection before the autonomous vehicle arrives at the next intersection according to the map data and the navigation information.

[0117] It can be understood that when determining the lane evaluation values of each candidate lane, in addition to considering the road conditions (traffic volume, obstacle distribution, etc.) of each candidate lane itself, the positional relationship between the candidate lane and the current lane where the autonomous vehicle is located, and the target lane corresponding to the next intersection before the autonomous vehicle arrives at the next intersection is also particularly important, for example, the closer the distance between the candidate lane and the current lane, the shorter the passing time required, the closer the distance between the candidate lane and the target lane, the fewer the number of lane changes required, the higher the safety of the autonomous vehicle driving, etc.​

[0118] Specifically, the autonomous vehicle can determine its driving direction corresponding to the current intersection according to the map data and the navigation information, and then take each lane of the direction road as a candidate lane, wherein the driving direction can include straight, left turn, right turn, etc. When determining the target lane corresponding to the next intersection before the ego vehicle reaches the next intersection, specifically, the driving direction when the ego vehicle reaches the next intersection can be determined, for example, when the driving direction when the ego vehicle reaches the next intersection is left turn, the target lane can be determined as the left turn lane or the lane closer to the left turn lane in the road; when the driving direction when the ego vehicle reaches the next intersection is right turn, the target lane can be determined as the right turn lane or the lane closer to the right turn lane in the road; when the driving direction when the ego vehicle reaches the next intersection is straight, the target lane can be determined as the straight lane in the road, and when there are multiple straight lanes, the closest straight lane can be selected according to the current lane of the vehicle, or the lane selection rule is preset, for example, the left lane or the right lane is determined as the target lane, which is all possible.

[0119] As shown in the intersection of FIG. 4B, the current lane of the ego vehicle 410 is lane A, the driving direction of the current intersection is straight, and the candidate lanes after passing through the intersection are lane A', lane B' and lane C'. When the driving direction when the ego vehicle 410 reaches the next intersection is left turn, the target lane can be determined as lane A'; when the driving direction when the ego vehicle 410 reaches the next intersection is right turn, the target lane can be determined as lane C'; and when the driving direction when the ego vehicle 410 reaches the next intersection is straight, the target lane can be determined as lane B'. Figure 4 As shown in the intersection of FIG. 5B, the current lane of the ego vehicle 510 is lane C, the driving direction of the current intersection is right turn, and the candidate lanes after passing through the intersection are lane A', lane B' and lane C'. The navigation information shows that the driving direction when reaching the next intersection is left turn, so the target lane corresponding to the next intersection before reaching the next intersection can be determined as lane A'.

[0120] Figure 5 As shown in the intersection of FIG. 5B, the current lane of the ego vehicle 510 is lane C, the driving direction of the current intersection is right turn, and the candidate lanes after passing through the intersection are lane A', lane B' and lane C'. The navigation information shows that the driving direction when reaching the next intersection is left turn, so the target lane corresponding to the next intersection before reaching the next intersection can be determined as lane A'.

[0121] S320: For each candidate lane, a first lane evaluation value from the current lane to the candidate lane is calculated according to the ego vehicle state information and the obstacle state information; and a second lane evaluation value from the candidate lane to the target lane is calculated according to the ego vehicle state information and the obstacle state information.

[0122] ​In the embodiments of the present application, when lane selection is performed, the efficiency and safety of traffic can be mainly considered, therefore, the first lane evaluation value can include a first traffic efficiency evaluation value and / or a first traffic safety evaluation value; and the second lane evaluation value can also include a second traffic efficiency evaluation value and / or a second traffic safety evaluation value. It can be understood that the traffic efficiency evaluation value can identify the driving efficiency of the ego vehicle, the higher the traffic efficiency evaluation value of any candidate lane, the higher the traffic efficiency of the ego vehicle, and the more suitable the candidate lane for the ego vehicle; and the traffic safety evaluation value identifies the driving safety of the ego vehicle, the higher the traffic safety evaluation value of any candidate lane, the higher the traffic safety of the ego vehicle, and the more suitable the candidate lane for the ego vehicle.

[0123] Specifically, the above obstacle state information can include static obstacle state information and dynamic obstacle state information, for each candidate lane, when calculating the first lane evaluation value from the current lane to the candidate lane, the smooth continuous curve length from the current lane to the candidate lane can be determined for each candidate lane, and based on the smooth continuous curve length, the first traffic efficiency evaluation value from the current lane to the candidate lane is determined; the first traffic efficiency evaluation value is negatively correlated with the smooth continuous curve length; for each candidate lane, the static structure risk coefficient corresponding to the candidate lane is calculated according to the static obstacle state information, and the dynamic traffic risk coefficient corresponding to the candidate lane is calculated according to the ego vehicle state information and the dynamic obstacle state information; and the first traffic safety evaluation value from the current lane to the candidate lane is calculated based on the static structure risk coefficient and the dynamic traffic risk coefficient; wherein the first traffic safety evaluation value is negatively correlated with the static structure risk coefficient and the dynamic traffic risk coefficient; the static structure risk coefficient is related to the relative position of the static obstacle and the lane and the probability of the static obstacle being observed; the dynamic traffic risk coefficient is related to the collision time of the ego vehicle and the dynamic obstacle; then the first lane evaluation value from the current lane to the candidate lane is calculated according to the first traffic efficiency evaluation value and the first traffic safety evaluation value; the first lane evaluation value is positively correlated with the first traffic efficiency evaluation value and the first traffic safety evaluation value.

[0124] For each candidate lane, when determining the smooth continuous curve length from the current lane to the candidate lane, the pose points of the current lane and the candidate lane can be first determined based on the map data, the pose points can be a vector including at least 4 parameters, respectively: x coordinate, y coordinate, heading angle and curvature value; and then the connecting line between the two pose points is determined based on any curve fitting method, and the length of the connecting line is the smooth continuous curve length from the current lane to the candidate lane. The above curve fitting method can include, for example, a third-order Bezier curve or a cubic spiral curve.

[0125] The smooth continuous curve length of the current lane to each candidate lane can identify the distance from the current lane to each candidate lane, and further, the distance can identify the travel time from the ego vehicle from the current lane to each candidate lane, and therefore, based on the calculated smooth continuous curve length, a first travel efficiency evaluation value of the current lane to each candidate lane can be determined, wherein the first travel efficiency evaluation value is negatively correlated with the smooth continuous curve length. That is, the greater the smooth continuous curve length of the current lane to any candidate lane, the longer the travel time required for the ego vehicle from the current lane to the candidate lane, and the lower the travel efficiency.

[0126] It can be understood that the relative position of the static obstacle and the lane, and the probability of the static obstacle being observed, are related to the lane safety, for example, the greater the relative position of the static obstacle and the lane, that is, the farther the distance between the static obstacle and the lane, the greater the lane safety; therefore, the above-mentioned static structure risk coefficient is related to the relative position of the static obstacle and the lane and the probability of the static obstacle being observed. The time to collision (TTC, Time to Collision) of the ego vehicle and the dynamic obstacle can reflect the relative position size of the ego vehicle and the dynamic obstacle, for example, the longer the collision time, the higher the ego vehicle travel safety, and therefore, the above-mentioned dynamic traffic risk coefficient is related to the collision time of the ego vehicle and the dynamic obstacle. And the greater the static structure risk coefficient and the dynamic traffic risk coefficient, the lower the ego vehicle travel safety, and therefore, the first travel safety evaluation value is negatively correlated with the static structure risk coefficient and the dynamic traffic risk coefficient.

[0127] When calculating the first lane evaluation value, for example, the sum of the first travel efficiency evaluation value and the first travel safety evaluation value corresponding to any candidate lane can be taken as the first lane evaluation value of the current lane to the candidate lane; or the weight of the first travel efficiency evaluation value and the first travel safety evaluation value can be set, and then the weighted result of the first travel efficiency evaluation value and the first travel safety evaluation value can be taken as the first lane evaluation value, which is all possible.

[0128] For each candidate lane, when calculating the second lane evaluation value from the candidate lane to the target lane, the minimum number of lane changes from the candidate lane to the target lane can be determined. Based on the vehicle's state information and obstacle state information, the expected travel time and expected lane change time for each lane traversed by the vehicle when traveling from the candidate lane to the target lane using the minimum number of lane changes are calculated. The expected travel time for any lane is related to the vehicle's expected travel distance in that lane and the obstacle distribution in that lane. The expected lane change time for any lane is positively correlated with the length of the lane's interchangeable section. Then, based on the minimum number of lane changes and each lane... The expected travel time and expected lane change time corresponding to each lane are used to calculate the second traffic efficiency assessment value from the candidate lane to the target lane. The second traffic efficiency assessment value is negatively correlated with the minimum number of lane changes, the expected travel time, and the expected lane change time. Then, based on the minimum number of lane changes, the second traffic safety assessment value from the candidate lane to the target lane is calculated. The second traffic safety assessment value is negatively correlated with the minimum number of lane changes. Finally, based on the second traffic efficiency assessment value and the second traffic safety assessment value, the second lane assessment value from the candidate lane to the target lane is calculated. The second lane assessment value is positively correlated with both the second traffic efficiency assessment value and the second traffic safety assessment value.

[0129] The minimum number of lane changes required to travel from each candidate lane to the target lane; that is, the minimum number of lane changes a vehicle needs to make to travel from that candidate lane to the target lane, without any lane reversal during the lane change process. For example... Figure 6 As shown, it displays the minimum number of lane changes for each candidate lane. When the target lane is lane a, the minimum number of lane changes for each candidate lane is as follows: the minimum number of lane changes for lane a is 0, the minimum number of lane changes for lane b is 1, the minimum number of lane changes for lane c is 2, the minimum number of lane changes for lane d is 3, and the minimum number of lane changes for lane d is 4.

[0130] It is understandable that many traffic accidents occur during lane changes, such as rear-end collisions caused by lane changes. Therefore, the number of lane changes directly affects the safety of vehicle driving. In this embodiment, the safety of each candidate lane can be measured based on the number of lane changes. Specifically, a second traffic safety assessment value from each candidate lane to the target lane can be calculated based on the minimum number of lane changes. The second traffic safety assessment value is negatively correlated with the minimum number of lane changes. That is, the smaller the minimum number of lane changes, the higher the vehicle traffic safety, and the larger the second traffic safety assessment value.

[0131] The obstacle distribution can include obstacle quantity, obstacle arrangement form, and obstacle flow, etc., and the lane-changing interval of any lane can include the length of the dashed line between the lane and the next lane. The fewer the minimum number of lane changes, the shorter the expected travel time, and the shorter the expected lane-changing time, the higher the travel efficiency of the ego vehicle. Therefore, the second travel efficiency evaluation value is negatively correlated with the minimum number of lane changes, the expected travel time, and the expected lane-changing time.

[0132] When calculating the second lane evaluation value, for example, the sum of the second travel efficiency evaluation value and the second travel safety evaluation value corresponding to any candidate lane can be taken as the second lane evaluation value of the candidate lane to the target lane; or the weights of the second travel efficiency evaluation value and the second travel safety evaluation value can be set, and then the weighted result of the second travel efficiency evaluation value and the second travel safety evaluation value can be taken as the second lane evaluation value, which is all possible.

[0133] S330: For each candidate lane, calculate the lane evaluation value of the candidate lane according to the first lane evaluation value and the second lane evaluation value.

[0134] Specifically, for each candidate lane, a first weight of the first lane evaluation value and a second weight of the second lane evaluation value can be determined; then the first lane evaluation value and the second lane evaluation value are weighted based on the first weight and the second weight, and the weighted result is taken as the lane evaluation value of the candidate lane. The first weight and the second weight can be set to the same value or different values, which is all possible, and the specific values are not limited by the embodiments of the application.

[0135] In one specific embodiment, the lane evaluation value F(N) of any candidate lane can be calculated according to the following formula: F(N) = F1(N) + F2(N).

[0136] F1(N) is used to measure the travel efficiency, and F2(N) is used to measure the travel safety.

[0137] F1(N) = G1(N) + H1(N) = g1(d) + h1(c) + h2(s, p1) + h3(p2)

[0138] G1(N) represents the efficiency cost of the current lane S to the candidate lane N, which is measured by the expected travel time, and G1(N) is negatively correlated with d, which is represented by g1(d), and d is the smooth continuous curve length from the current lane S to the candidate lane N.

[0139] H1(N) represents the efficiency cost of the candidate lane N to the target lane E, measured by the expected travel time. Wherein, H1(N) is negatively correlated with c, wherein c is the minimum number of lane changes from the candidate lane N to the target lane E, denoted by h1(c); at the same time, H1(N) is correlated with the expected travel time t(n) on each lane n from the candidate lane N to the target lane E, through the shortest number of lane changes c, in the direction of the target lane, and the expected travel time t(n) is a function of the expected driving length s on the lane n and the obstacle distribution p1 on the lane n, that is: t(n) = h2(s, p1), and the expected travel time is negatively correlated with the distribution p2 of the lane changeable interval (the length of the virtual line between the lanes) on the lane, that is h3(p2).

[0140] F2(N) = G2(N) + H2(N) = g2(r1, r2) + h2(c)

[0141] G2(N) represents the safety cost of the current lane S to the candidate lane N, measured by the expected collision probability, and G2(N) is correlated with the road static structure risk coefficient r1 and the dynamic traffic risk coefficient r2 from S to N, denoted by g2(r1, r2).

[0142] H2(N) represents the safety cost of the candidate lane N to the target lane E, measured by the expected collision or expected collision risk probability, and H2(N) is negatively correlated with c, wherein c is the minimum number of lane changes from the candidate lane N to the target lane E, denoted by h2(c).

[0143] In another implementation, the lane evaluation value of each candidate lane can also be determined based on a pre-trained network model. Specifically, a certain amount of expert driving data can be collected in advance, and the expert driving data includes intersection driving data, that is, the corresponding exit lane selection result of the autonomous vehicle passing through the intersection. The network model trained based on the expert driving data can perform intersection lane selection.

[0144] When performing intersection lane selection, the obtained map data, navigation information, vehicle state information and obstacle state information can be input into the pre-trained lane selection model, and the lane selection model can output the probability value of each candidate lane corresponding to the vehicle driving out of the intersection; the size of the probability value of each candidate lane indicates the probability of each candidate lane being suitable for passing, and then the probability value of each candidate lane can be used as the lane evaluation value of the corresponding candidate lane.

[0145] After determining the lane evaluation value of each candidate lane, in one implementation, the lane evaluation value of each candidate lane can be directly used as its passing evaluation value.

[0146] It can be understood that, in determining the traffic evaluation value of each candidate lane, in addition to the lane condition of each candidate lane itself, there can be other factors that will affect the traffic condition of the lane. Therefore, in an embodiment, before determining the traffic evaluation value of each candidate lane based on the lane evaluation value of the candidate lane, a safety evaluation value of each candidate lane can also be calculated; wherein the safety evaluation value can include: a trajectory safety evaluation result, and / or, intersection collision risk; and then the traffic evaluation value of each candidate lane can be determined based on the lane evaluation value and the safety evaluation value of the candidate lane.

[0147] In calculating the trajectory safety evaluation result of each candidate lane, a planned trajectory from the current lane of the ego vehicle to the candidate lane can be generated for each candidate lane, and the trajectory safety evaluation result corresponding to the planned trajectory can be calculated. For example, based on the current trajectory planning module of the autonomous vehicle, a planned trajectory from the current lane of the ego vehicle to each candidate lane can be generated, and the corresponding trajectory safety evaluation result can be calculated by comprehensively considering the safety, smoothness, etc. of the planned trajectory.

[0148] It can be understood that, due to the differences in complexity, intersection structure, etc., different intersections have different safety. In the embodiments of the present application, the intersection collision risk value corresponding to each intersection can be determined in advance according to the historical vehicle driving safety situation, etc. and stored, for example, each intersection can be numbered, and the intersection number and the corresponding collision risk value can be stored correspondingly. Thus, when selecting the intersection lane, the intersection collision risk value corresponding to each intersection stored in advance can be obtained, and the intersection collision risk corresponding to the current intersection can be determined.

[0149] The greater the trajectory safety evaluation result, the higher the safety of the candidate lane, and the more suitable for the ego vehicle to pass through; and the greater the intersection collision risk, the lower the safety of the candidate lane, and the less suitable for the ego vehicle to pass through; therefore, the safety evaluation value of each candidate lane is positively correlated with the trajectory safety evaluation result and negatively correlated with the intersection collision risk. In an implementation, for example, the difference between the trajectory safety evaluation result and the intersection collision risk of any candidate lane can be taken as the safety evaluation value of the candidate lane.

[0150] In determining the traffic evaluation value of each candidate lane, the sum of the lane evaluation value and the safety evaluation value of each candidate lane can be taken as the traffic evaluation value of the candidate lane.

[0151] By comprehensively considering the lane evaluation value and the safety evaluation value of each candidate lane to determine the traffic evaluation value of each candidate lane, each candidate lane can be more comprehensively evaluated, and thus the accuracy of lane selection can be improved when selecting the lane.

[0152] S230: Based on the current trajectory planning algorithm and control tracking algorithm, determine the control cost of each candidate lane.

[0153] As mentioned above, considering only the intersections of candidate lanes may result in excessively large subsequent control tracking errors, leading to traffic accidents. Therefore, in this embodiment, to ensure the safety of autonomous vehicles, the control costs of each candidate lane can be comprehensively considered after determining the traffic evaluation value of each candidate lane.

[0154] Specifically, the control cost of each candidate lane can be determined based on the current trajectory planning algorithm and control tracking algorithm. In other words, for each candidate lane, the trajectory error corresponding to the vehicle's movement from the current lane to that candidate lane can be calculated, thereby filtering out candidate lanes that do not meet the vehicle's driving safety requirements.

[0155] In one implementation, the first historical lateral error of the trajectory planning algorithm and the second historical lateral error of the control tracking algorithm can be statistically analyzed. Then, based on the first and second historical lateral errors, the control cost of each candidate lane is calculated, where the control cost is positively correlated with both the first and second historical lateral errors. For example, the sum of the first and second historical lateral errors can be determined as the control cost of each candidate lane.

[0156] After lane selection, the autonomous vehicle needs to control its movement based on its current trajectory planning and control tracking algorithms. Errors in these algorithms directly affect whether each candidate lane allows the vehicle to pass safely. For example, if the control cost of any candidate lane is too high, such as... Figure 1 As shown, when the leftmost lane is designated as the exit lane, the actual vehicle trajectory 140 will have too small a distance between itself and the static obstacle at the road edge, potentially leading to a dangerous accident. Therefore, the first historical lateral error of the trajectory planning algorithm and the second historical lateral error of the control tracking algorithm can accurately reflect the current computing power of the trajectory planning algorithm and the control tracking algorithm. By calculating the control cost of each candidate lane based on the first and second historical lateral errors, each candidate lane can be accurately evaluated.

[0157] When calculating the control cost of each candidate lane based on the first historical lateral error and the second historical lateral error, in one implementation, the mean and variance of the error can be determined based on the distribution of trajectory planning and control tracking errors from historical data, and then the control cost can be determined based on the determined mean and variance of the error.

[0158] In order to more accurately calculate the regulation and control cost of each candidate lane, in another implementation manner, for each candidate lane, a plurality of candidate trajectories from the current lane of the ego vehicle to the candidate lane can be generated based on a trajectory planning algorithm, for each candidate trajectory, a planning cost of the candidate trajectory can be calculated, and a control tracking algorithm can be used to control and track the candidate trajectory to obtain a corresponding control error, and the tracking cost of the candidate trajectory can be calculated according to the control error; the tracking cost is positively correlated with the control error; for each candidate trajectory, the trajectory cost of the candidate trajectory can be calculated according to the planning cost and the tracking cost of the candidate trajectory; the trajectory cost is negatively correlated with the planning cost and positively correlated with the tracking cost; the candidate trajectory with the lowest trajectory cost is determined, and the trajectory cost of the candidate trajectory is taken as the regulation and control cost of the candidate lane.

[0159] When calculating the planning cost of each candidate trajectory, the implementation environment can be at the vehicle end or in the cloud. The specific calculation method can be based on a method considering safety, comfort, efficiency and other factors, or a trajectory evaluation network can be used to give the planning cost of the candidate trajectory.

[0160] The control tracking of the candidate trajectory can obtain the control error of the candidate trajectory, and the specific method can be: constructing a vehicle dynamics model, and performing control pre-tracking based on the vehicle dynamics model; or predicting the control error based on a neural network, wherein the input of the neural network is the candidate trajectory, and the output is the corresponding control error. The above method can be implemented at the vehicle end or in the cloud, wherein the cloud can use sufficient computing resources to perform fast and accurate simulation calculation or model reasoning, and the calculation result can be fed back to the vehicle end in real time through the 5G network.

[0161] That is, for each candidate lane, a plurality of candidate trajectories can be generated, and the planning cost of each candidate trajectory can be calculated, which can indicate the performance (such as smoothness, safety, etc.) of the corresponding candidate trajectory. The greater the planning cost, the worse the performance of the candidate trajectory, and the smaller the planning cost, the better the performance of the candidate trajectory. Moreover, the control tracking of each candidate trajectory can be performed to calculate the corresponding control error, and the control error can indicate the trackability of the corresponding candidate trajectory. The greater the control error, the worse the trackability of the candidate trajectory, and the smaller the control error, the stronger the trackability of the candidate trajectory. Finally, the trajectory cost of the candidate trajectory can be calculated according to the planning cost and the tracking cost of the candidate trajectory. The trajectory cost is positively correlated with the planning cost and the tracking cost, for example, the sum of the tracking cost and the trajectory cost can be determined as the regulation and control cost of each candidate lane. The candidate trajectory with the lowest trajectory cost, that is, the trajectory with good performance and strong trackability, is determined, and the trajectory cost of the candidate trajectory is taken as the regulation and control cost of the candidate lane.

[0162] The planning cost of each candidate trajectory is calculated by comprehensively considering the safety and smoothness of the candidate trajectory, and a trajectory safety evaluation result is calculated as the planning cost of the candidate trajectory.

[0163] In S240, the optimal lane is determined from the candidate lanes according to the passing evaluation value and the regulation and control cost of each candidate lane.

[0164] The passing evaluation value of each candidate lane indicates the degree to which the candidate lane is suitable for vehicle passing, and the larger the passing evaluation value, the more suitable the candidate lane is for the ego vehicle to pass; and the regulation and control cost of the candidate lane indicates the total error size of the trajectory planning and control tracking when the ego vehicle passes the intersection through the candidate lane, and the larger the regulation and control cost, the larger the total error.

[0165] In the embodiments of the present application, after the passing evaluation value and the regulation and control cost of each candidate lane are calculated, the optimal lane can be determined from the candidate lanes according to the passing evaluation value and the regulation and control cost of each candidate lane. For example, candidate lanes whose regulation and control cost meet the preset requirements can be first determined as candidate lanes from the candidate lanes; and then the candidate lane with the highest passing evaluation value can be determined as the optimal lane from the candidate lanes.

[0166] The preset requirements may, for example, include that the ego vehicle will not collide with static obstacles when passing the intersection through the candidate lane under the influence of the regulation and control cost of the candidate lane. After the candidate lanes are screened according to the regulation and control cost, the candidate lanes determined are lanes that can ensure the safe passing of the ego vehicle, and further, the candidate lane with the highest passing evaluation value, that is, the lane with the highest safety and the highest passing efficiency, is determined as the optimal lane, so that the safety and the passing efficiency of the ego vehicle can be considered, and the accuracy of lane determination can be improved.

[0167] In the embodiments of the present application, when selecting the intersection lane, not only the lane evaluation value of each candidate lane, that is, the current road condition of each candidate lane, but also the regulation and control cost of each candidate lane, that is, the corresponding planning and control result when each candidate lane is used as the intersection exit lane, can be considered based on the current trajectory planning algorithm and control tracking algorithm. Therefore, the optimal lane can be selected based on the lane evaluation value and the regulation and control cost of each candidate lane, so that the finally determined intersection exit lane can be optimal and can meet the current planning and control algorithm power, and the situation that the autonomous vehicle collides with obstacles such as road edges, fences, and barrels during passing the intersection can be avoided, so that the accuracy of the intersection lane selection of the autonomous vehicle can be improved, and the driving safety of the autonomous vehicle can be ensured.

[0168] Figure 7A structural schematic diagram of a lane selection device for an intersection is shown, the device comprises:

[0169] The data acquisition module 710 is configured to acquire map data corresponding to a current intersection, and navigation information, vehicle state information and obstacle state information;

[0170] The passing evaluation value determination module 720 is configured to determine lane evaluation values of each candidate lane for the vehicle to drive out of the intersection based on the map data, the navigation information, the vehicle state information and the obstacle state information, and determine, for each candidate lane, a passing evaluation value of the candidate lane based on the lane evaluation value of the candidate lane;

[0171] The regulation and control cost determination module 730 is configured to determine regulation and control costs of each candidate lane based on a current trajectory planning algorithm and a control tracking algorithm;

[0172] The optimal lane determination module 740 is configured to determine an optimal lane from each candidate lane according to the passing evaluation value and the regulation and control cost of each candidate lane.

[0173] Optionally, the device further comprises:

[0174] The safety evaluation value calculation module is configured to calculate a safety evaluation value of each candidate lane; the safety evaluation value comprises a trajectory safety evaluation result and / or an intersection collision risk;

[0175] The passing evaluation value determination module 720 is specifically configured to:

[0176] For each candidate lane, determine a passing evaluation value of the candidate lane based on the lane evaluation value and the safety evaluation value of the candidate lane.

[0177] Optionally, when the safety evaluation value comprises a trajectory safety evaluation result and an intersection collision risk, the safety evaluation value calculation module is specifically configured to:

[0178] For each candidate lane, generate a planned trajectory from a current lane where the vehicle is located to the candidate lane, and calculate a trajectory safety evaluation result corresponding to the planned trajectory;

[0179] Acquire intersection collision risk values corresponding to each intersection that are pre-stored, and determine an intersection collision risk corresponding to the current intersection according to the intersection collision risk values corresponding to each intersection.

[0180] Optionally, the passing evaluation value determination module 720 comprises:

[0181] The lane determination sub-module is configured to determine a current lane in which the ego vehicle is located, and determine, according to the map data and the navigation information, each candidate lane corresponding to the ego vehicle driving out of the intersection, and a target lane corresponding to the ego vehicle before reaching a next intersection;

[0182] The lane evaluation value calculation sub-module is configured to, for each candidate lane, calculate, according to the ego vehicle state information and the obstacle state information, a first lane evaluation value from the current lane to the candidate lane, and calculate, according to the ego vehicle state information and the obstacle state information, a second lane evaluation value from the candidate lane to the target lane; the first lane evaluation value includes a first traffic efficiency evaluation value and / or a first traffic safety evaluation value; and the second lane evaluation value includes a second traffic efficiency evaluation value and / or a second traffic safety evaluation value.

[0183] The lane evaluation value combination sub-module is configured to, for each candidate lane, calculate, according to the first lane evaluation value and the second lane evaluation value, a lane evaluation value of the candidate lane.

[0184] Optionally, the obstacle state information includes static obstacle state information and dynamic obstacle state information; when the first lane evaluation value includes a first traffic efficiency evaluation value and a first traffic safety evaluation value, the lane evaluation value calculation sub-module is specifically configured to:

[0185] For each candidate lane, determine a smooth continuous curve length from the current lane to the candidate lane, and determine, based on the smooth continuous curve length, a first traffic efficiency evaluation value from the current lane to the candidate lane; the first traffic efficiency evaluation value is negatively correlated with the smooth continuous curve length.

[0186] For each candidate lane, calculate, according to the static obstacle state information, a static structure risk coefficient corresponding to the candidate lane, calculate, according to the ego vehicle state information and the dynamic obstacle state information, a dynamic traffic flow risk coefficient corresponding to the candidate lane, and calculate, based on the static structure risk coefficient and the dynamic traffic flow risk coefficient, a first traffic safety evaluation value from the current lane to the candidate lane; the first traffic safety evaluation value is negatively correlated with the static structure risk coefficient and the dynamic traffic flow risk coefficient; the static structure risk coefficient is negatively correlated with a relative position between a static obstacle and a lane and a probability of the static obstacle being observed; and the dynamic traffic flow risk coefficient is negatively correlated with a collision time between the ego vehicle and a dynamic obstacle.

[0187] Calculate, according to the first traffic efficiency evaluation value and the first traffic safety evaluation value, a first lane evaluation value from the current lane to the candidate lane; the first lane evaluation value is positively correlated with the first traffic efficiency evaluation value and the first traffic safety evaluation value.

[0188] Optionally, when the second lane evaluation value comprises a second passing efficiency evaluation value and a second passing safety evaluation value, the lane evaluation value calculation submodule is specifically configured to:

[0189] determine a minimum number of lane changes from the candidate lane to the target lane, and calculate, according to the ego vehicle state information and the obstacle state information, an expected passing time and an expected lane changing time corresponding to each lane passed by the ego vehicle from the candidate lane to the target lane through the minimum number of lane changes; the expected passing time corresponding to any lane is related to an expected driving distance of the ego vehicle in the lane and an obstacle distribution of the lane; the expected lane changing time corresponding to any lane is positively related to a lane changing interval length of the lane;

[0190] calculate a second passing efficiency evaluation value from the candidate lane to the target lane according to the minimum number of lane changes, the expected passing time and the expected lane changing time corresponding to each lane; the second passing efficiency evaluation value is negatively related to the minimum number of lane changes, the expected passing time and the expected lane changing time;

[0191] calculate a second passing safety evaluation value from the candidate lane to the target lane according to the minimum number of lane changes; the second passing safety evaluation value is negatively related to the minimum number of lane changes;

[0192] calculate a second lane evaluation value from the candidate lane to the target lane according to the second passing efficiency evaluation value and the second passing safety evaluation value; the second lane evaluation value is positively related to the second passing efficiency evaluation value and the second passing safety evaluation value.

[0193] Optionally, the lane evaluation value combination submodule is specifically configured to:

[0194] for each candidate lane, determine a first weight of the first lane evaluation value and a second weight of the second lane evaluation value;

[0195] weight the first lane evaluation value and the second lane evaluation value based on the first weight and the second weight, and take the weighting result as a lane evaluation value of the candidate lane.

[0196] Optionally, the passing evaluation value determination module 720 is specifically configured to:

[0197] input the map data, the navigation information, the ego vehicle state information and the obstacle state information into a lane selection model pre-trained to obtain probability values of each candidate lane corresponding to the ego vehicle driving out of the intersection; the lane selection model is pre-trained according to expert driving data;

[0198] The probability value of each candidate lane is taken as a lane evaluation value of the corresponding candidate lane.

[0199] Optionally, the regulation and control cost determination module 730 is specifically configured to:

[0200] statistically determine a first historical lateral error of the trajectory planning algorithm and a second historical lateral error of the control tracking algorithm;

[0201] According to the first historical lateral error and the second historical lateral error, the regulation and control cost of each candidate lane is calculated, and the regulation and control cost is positively correlated with the first historical lateral error and the second historical lateral error.

[0202] Optionally, the regulation and control cost determination module 730 is specifically configured to:

[0203] For each candidate lane, a plurality of candidate trajectories from a current lane where the ego vehicle is located to the candidate lane are generated based on the trajectory planning algorithm;

[0204] For each candidate trajectory, a planning cost of the candidate trajectory is calculated, and the control tracking algorithm is used to control and track the candidate trajectory to obtain a corresponding control error, and a tracking cost of the candidate trajectory is calculated according to the control error; the tracking cost is positively correlated with the control error;

[0205] For each candidate trajectory, a trajectory cost of the candidate trajectory is calculated according to the planning cost and the tracking cost of the candidate trajectory; the trajectory cost is negatively correlated with the planning cost and positively correlated with the tracking cost;

[0206] The candidate trajectory with the lowest trajectory cost is determined, and the trajectory cost of the candidate trajectory is taken as the regulation and control cost of the candidate lane.

[0207] Optionally, the optimal lane determination module 740 is specifically configured to:

[0208] Among the candidate lanes, a candidate lane with a regulation and control cost meeting a preset requirement is determined as a candidate lane;

[0209] Among the candidate lanes, a candidate lane with the highest passing evaluation value is determined as the optimal lane.

[0210] In the embodiments of the present application, when the intersection lane is selected, not only the lane evaluation value of each candidate lane, that is, the current road condition of each candidate lane, but also the regulation and control cost of each candidate lane, that is, the corresponding planning and control result if each candidate lane is used as the out-of-intersection lane, is considered based on the current trajectory planning algorithm and control tracking algorithm. Therefore, the optimal lane selection based on the lane evaluation value and the regulation and control cost of each candidate lane can ensure that the out-of-intersection lane determined finally is in a relatively optimal condition and can meet the computing power of the current planning and control algorithm, so as to avoid the collision between the autonomous vehicle and the obstacles such as the road edge, fence, cone barrel and the like during the passing of the intersection, thereby improving the accuracy of the intersection lane selection of the autonomous vehicle and ensuring the driving safety of the autonomous vehicle.

[0211] The device embodiments correspond to the method embodiments and have the same technical effects. For details, refer to the method embodiments. The device embodiments are based on the method embodiments. For details, refer to the method embodiments, which will not be described herein.

[0212] Next, a computer device provided by the embodiments of the present application is introduced. Please refer to Figure 8 , Figure 8 The computer device provided by the embodiments of the present application has a structure diagram as shown in the figure. The computer device includes:

[0213] one or more processors 40;

[0214] The processor 40 is coupled with a storage device 41, and the storage device 41 is used to store one or more programs,

[0215] When the one or more programs are executed by the one or more processors 40, the electronic device implements the technical solutions of the intersection lane selection method as described above. Figures 2 to 6

[0216] The embodiments of the present application also provide a computer readable storage medium, which stores a computer program, and the program is executed by a processor to implement the technical solutions of the intersection lane selection method as described above. Figures 2 to 6

[0217] The present application provides a computer program product, which includes a computer program, and the computer program is executed by a processor to implement the technical solutions of the intersection lane selection method as described above. Figures 2 to 6

[0218] Those skilled in the art can understand that the drawings are only schematic diagrams of the embodiments, and the modules or processes in the drawings are not necessarily required to implement the present application. ​​​

[0219] Those skilled in the art can understand that the modules in the device in the embodiments can be distributed in the device in the embodiments as described in the embodiments, or can be changed to be located in one or more devices different from the embodiments. The modules in the above embodiments can be combined into one module, or can be further split into multiple sub-modules.

[0220] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present application, and not to limit them; although the present application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that they can still modify the technical solutions recorded in the foregoing embodiments, or make equivalent replacements for some of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A method for selecting lanes at an intersection, characterized in that, The method includes: Obtain map data corresponding to the current intersection, as well as navigation information, vehicle status information, and obstacle status information; Based on the map data, the navigation information, the vehicle status information, and the obstacle status information, the lane evaluation value of each candidate lane corresponding to the vehicle exiting the intersection is determined, and for each candidate lane, the passage evaluation value of the candidate lane is determined based on the lane evaluation value of the candidate lane. Based on the current trajectory planning algorithm and control tracking algorithm, the control cost of each candidate lane is determined; The optimal lane is determined from among the candidate lanes based on the traffic assessment value and the control cost of each candidate lane.

2. The method according to claim 1, characterized in that, Before determining the traffic assessment value of each candidate lane based on its lane assessment value, the method further includes: Calculate a safety assessment value for each of the candidate lanes; the safety assessment value includes: trajectory safety assessment result, and / or, intersection collision risk; The step of determining the traffic assessment value of each candidate lane based on its lane assessment value includes: For each candidate lane, a traffic assessment value is determined based on the lane assessment value and safety assessment value of the candidate lane.

3. The method according to claim 2, characterized in that, When the safety assessment value includes the trajectory safety assessment result and the intersection collision risk, the step of calculating the safety assessment value for each candidate lane includes: For each candidate lane, a planned trajectory from the vehicle's current lane to that candidate lane is generated, and the trajectory safety assessment result corresponding to the planned trajectory is calculated; Obtain the pre-stored intersection collision risk values ​​corresponding to each intersection, and determine the intersection collision risk corresponding to the current intersection based on the intersection collision risk values ​​corresponding to each intersection.

4. The method according to claim 1, characterized in that, The step of determining the lane evaluation value of each candidate lane corresponding to the vehicle exiting the intersection based on the map data, the navigation information, the vehicle status information, and the obstacle status information includes: The system determines the current lane of the vehicle and, based on the map data and navigation information, determines the candidate lanes corresponding to the vehicle's exit from the intersection, as well as the target lane corresponding to the vehicle's arrival at the next intersection. For each candidate lane, a first lane evaluation value from the current lane to the candidate lane is calculated based on the vehicle status information and the obstacle status information; and a second lane evaluation value from the candidate lane to the target lane is calculated based on the vehicle status information and the obstacle status information; the first lane evaluation value includes a first traffic efficiency evaluation value and / or a first traffic safety evaluation value; the second lane evaluation value includes a second traffic efficiency evaluation value and / or a second traffic safety evaluation value. For each candidate lane, a lane evaluation value is calculated based on the first lane evaluation value and the second lane evaluation value.

5. The method according to claim 4, characterized in that, The obstacle state information includes static obstacle state information and dynamic obstacle state information; when the first lane evaluation value includes a first traffic efficiency evaluation value and a first traffic safety evaluation value, the step of calculating the first lane evaluation value from the current lane to the candidate lane based on the vehicle state information and the obstacle state information for each candidate lane includes: For each candidate lane, the length of a smooth, continuous curve from the current lane to the candidate lane is determined, and based on the length of the smooth, continuous curve, a first traffic efficiency assessment value from the current lane to the candidate lane is determined; the first traffic efficiency assessment value is negatively correlated with the length of the smooth, continuous curve. For each candidate lane, a static structural risk coefficient is calculated based on the static obstacle state information, and a dynamic traffic flow risk coefficient is calculated based on the vehicle state information and the dynamic obstacle state information. A first traffic safety assessment value is then calculated from the current lane to the candidate lane based on the static structural risk coefficient and the dynamic traffic flow risk coefficient. The first traffic safety assessment value is negatively correlated with both the static structural risk coefficient and the dynamic traffic flow risk coefficient. The static structural risk coefficient is related to the relative position of the static obstacle and the lane, as well as the probability of the static obstacle being observed. The dynamic traffic flow risk coefficient is related to the collision time between the vehicle and the dynamic obstacle. Based on the first traffic efficiency assessment value and the first traffic safety assessment value, a first lane assessment value is calculated from the current lane to the candidate lane; the first lane assessment value is positively correlated with both the first traffic efficiency assessment value and the first traffic safety assessment value.

6. The method according to claim 4, characterized in that, When the second lane evaluation value includes a second traffic efficiency evaluation value and a second traffic safety evaluation value, the step of calculating the second lane evaluation value from the candidate lane to the target lane based on the vehicle status information and the obstacle status information includes: The minimum number of lane changes required to reach the target lane from the candidate lane is determined. Based on the vehicle status information and the obstacle status information, the expected travel time and expected lane change time for each lane traversed by the vehicle from the candidate lane to the target lane using the minimum number of lane changes are calculated. The expected travel time for any lane is related to the expected travel distance of the vehicle in that lane and the obstacle distribution in that lane. The expected lane change time for any lane is positively correlated with the length of the lane's interchangeable section. Based on the minimum number of lane changes, the expected travel time and expected lane change time for each lane, a second traffic efficiency assessment value is calculated from the candidate lane to the target lane; the second traffic efficiency assessment value is negatively correlated with the minimum number of lane changes, the expected travel time, and the expected lane change time. A second traffic safety assessment value for the candidate lane to the target lane is calculated based on the minimum number of lane changes; the second traffic safety assessment value is negatively correlated with the minimum number of lane changes. Based on the second traffic efficiency assessment value and the second traffic safety assessment value, a second lane assessment value is calculated from the candidate lane to the target lane; the second lane assessment value is positively correlated with both the second traffic efficiency assessment value and the second traffic safety assessment value.

7. The method according to claim 4, characterized in that, The step of calculating the lane evaluation value of each candidate lane based on the first lane evaluation value and the second lane evaluation value includes: For each of the candidate lanes, a first weight for the first lane evaluation value and a second weight for the second lane evaluation value are determined; Based on the first weight and the second weight, the first lane evaluation value and the second lane evaluation value are weighted, and the weighted result is used as the lane evaluation value of the candidate lane.

8. The method according to claim 1, characterized in that, The step of determining the lane evaluation value of each candidate lane corresponding to the vehicle exiting the intersection based on the map data, the navigation information, the vehicle status information, and the obstacle status information includes: The map data, navigation information, vehicle status information, and obstacle status information are input into a pre-trained lane selection model to obtain the probability values ​​of each candidate lane corresponding to the vehicle exiting the intersection; the lane selection model is pre-trained based on expert driving data. The probability value of each candidate lane is used as the lane evaluation value for that candidate lane.

9. The method according to any one of claims 1-8, characterized in that, The step of determining the control cost of each candidate lane based on the current trajectory planning algorithm and control tracking algorithm includes: The first historical lateral error of the trajectory planning algorithm and the second historical lateral error of the control tracking algorithm are statistically analyzed. Based on the first historical lateral error and the second historical lateral error, the regulation cost of each candidate lane is calculated, and the regulation cost is positively correlated with both the first historical lateral error and the second historical lateral error.

10. The method according to any one of claims 1-8, characterized in that, The step of determining the control cost of each candidate lane based on the current trajectory planning algorithm and control tracking algorithm includes: For each candidate lane, multiple candidate trajectories from the vehicle's current lane to that candidate lane are generated based on the trajectory planning algorithm; For each candidate trajectory, the planning cost of the candidate trajectory is calculated, and the control tracking algorithm is used to control and track the candidate trajectory to obtain the corresponding control error. The tracking cost of the candidate trajectory is calculated based on the control error; the tracking cost is positively correlated with the control error. For each candidate trajectory, the trajectory cost is calculated based on the planning cost and the tracking cost of the candidate trajectory; the trajectory cost is positively correlated with the planning cost and the tracking cost. The candidate trajectory with the lowest trajectory cost is determined, and the trajectory cost of the candidate trajectory is used as the control cost of the candidate lane.

11. The method according to any one of claims 1-8, characterized in that, The step of determining the optimal lane among the candidate lanes based on the traffic assessment value and regulation cost of each candidate lane includes: Among the candidate lanes, the candidate lanes whose control costs meet the preset requirements are determined as alternative lanes; Among the candidate lanes, the lane with the highest traffic assessment value is determined as the optimal lane.

12. A lane selection device for an intersection, characterized in that, The device includes: The data acquisition module is used to acquire map data corresponding to the current intersection, as well as navigation information, vehicle status information, and obstacle status information; The traffic assessment value determination module is used to determine the lane assessment value of each candidate lane corresponding to the exit of the intersection based on the map data, the navigation information, the vehicle status information and the obstacle status information, and to determine the traffic assessment value of each candidate lane based on the lane assessment value of the candidate lane. The regulation cost determination module is used to determine the regulation cost of each candidate lane based on the current trajectory planning algorithm and control tracking algorithm. The optimal lane determination module is used to determine the optimal lane among the candidate lanes based on the traffic assessment value and the control cost of each candidate lane.

13. The apparatus according to claim 12, characterized in that, The device further includes: A safety assessment value calculation module is used to calculate the safety assessment value for each of the candidate lanes; the safety assessment value includes: trajectory safety assessment result, and / or, intersection collision risk; The passage assessment value determination module is specifically used for: For each candidate lane, a traffic assessment value is determined based on the lane assessment value and safety assessment value of the candidate lane.

14. The apparatus according to claim 13, characterized in that, When the safety assessment value includes trajectory safety assessment results and intersection collision risk, the safety assessment value calculation module is specifically used for: For each candidate lane, a planned trajectory from the vehicle's current lane to that candidate lane is generated, and the trajectory safety assessment result corresponding to the planned trajectory is calculated; Obtain the pre-stored intersection collision risk values ​​corresponding to each intersection, and determine the intersection collision risk corresponding to the current intersection based on the intersection collision risk values ​​corresponding to each intersection.

15. The apparatus according to claim 12, characterized in that, The passage assessment value determination module includes: The lane determination submodule is used to determine the current lane of the vehicle, and based on the map data and the navigation information, determine the candidate lanes corresponding to the exit of the vehicle from the intersection, as well as the target lane corresponding to the vehicle before reaching the next intersection. The lane evaluation value calculation submodule is used to calculate, for each candidate lane, a first lane evaluation value from the current lane to the candidate lane based on the vehicle status information and the obstacle status information; and to calculate a second lane evaluation value from the candidate lane to the target lane based on the vehicle status information and the obstacle status information; the first lane evaluation value includes a first traffic efficiency evaluation value and / or a first traffic safety evaluation value; the second lane evaluation value includes a second traffic efficiency evaluation value and / or a second traffic safety evaluation value; The lane evaluation value combination submodule is used to calculate the lane evaluation value of each candidate lane based on the first lane evaluation value and the second lane evaluation value.

16. The apparatus according to claim 15, characterized in that, The obstacle status information includes static obstacle status information and dynamic obstacle status information; when the first lane evaluation value includes a first traffic efficiency evaluation value and a first traffic safety evaluation value, the lane evaluation value calculation submodule is specifically used for: For each candidate lane, the length of a smooth, continuous curve from the current lane to the candidate lane is determined, and based on the length of the smooth, continuous curve, a first traffic efficiency assessment value from the current lane to the candidate lane is determined; The first traffic efficiency assessment value is negatively correlated with the length of the smooth continuous curve; For each candidate lane, the static structure risk coefficient corresponding to the candidate lane is calculated based on the static obstacle state information, and the dynamic traffic flow risk coefficient corresponding to the candidate lane is calculated based on the vehicle state information and the dynamic obstacle state information. Based on the static structural risk coefficient and the dynamic traffic flow risk coefficient, a first traffic safety assessment value from the current lane to the candidate lane is calculated; wherein, the first traffic safety assessment value is negatively correlated with the static structural risk coefficient and the dynamic traffic flow risk coefficient; the static structural risk coefficient is related to the relative position of the static obstacle and the lane and the probability that the static obstacle is observed; the dynamic traffic flow risk coefficient is related to the collision time between the vehicle and the dynamic obstacle; Based on the first traffic efficiency assessment value and the first traffic safety assessment value, a first lane assessment value is calculated from the current lane to the candidate lane; the first lane assessment value is positively correlated with both the first traffic efficiency assessment value and the first traffic safety assessment value.

17. The apparatus according to claim 15, characterized in that, When the second lane assessment value includes a second traffic efficiency assessment value and a second traffic safety assessment value, the lane assessment value calculation submodule is specifically used for: The minimum number of lane changes required to reach the target lane from the candidate lane is determined. Based on the vehicle status information and the obstacle status information, the expected travel time and expected lane change time for each lane traversed by the vehicle from the candidate lane to the target lane using the minimum number of lane changes are calculated. The expected travel time for any lane is related to the expected travel distance of the vehicle in that lane and the obstacle distribution in that lane. The expected lane change time for any lane is positively correlated with the length of the lane's interchangeable section. Based on the minimum number of lane changes, the expected travel time and expected lane change time for each lane, a second traffic efficiency assessment value is calculated from the candidate lane to the target lane; the second traffic efficiency assessment value is negatively correlated with the minimum number of lane changes, the expected travel time, and the expected lane change time. A second traffic safety assessment value for the candidate lane to the target lane is calculated based on the minimum number of lane changes; the second traffic safety assessment value is negatively correlated with the minimum number of lane changes. Based on the second traffic efficiency assessment value and the second traffic safety assessment value, a second lane assessment value is calculated from the candidate lane to the target lane; the second lane assessment value is positively correlated with both the second traffic efficiency assessment value and the second traffic safety assessment value.

18. The apparatus according to any one of claims 12-17, characterized in that, The regulatory cost determination module is specifically used for: For each candidate lane, multiple candidate trajectories from the vehicle's current lane to that candidate lane are generated based on the trajectory planning algorithm; For each candidate trajectory, the planning cost of the candidate trajectory is calculated, and the control tracking algorithm is used to control and track the candidate trajectory to obtain the corresponding control error. The tracking cost of the candidate trajectory is then calculated based on the control error. The tracking cost is positively correlated with the control error; For each candidate trajectory, the trajectory cost of the candidate trajectory is calculated based on the planning cost and tracking cost of the candidate trajectory; The trajectory cost is positively correlated with the planning cost and the tracking cost; The candidate trajectory with the lowest trajectory cost is determined, and the trajectory cost of the candidate trajectory is used as the control cost of the candidate lane.

19. A computer device, characterized in that, include: The memory and the processor are coupled; The memory is used to store one or more computer instructions; The processor is used to execute one or more computer instructions to implement the intersection lane selection method as described in any one of claims 1 to 11.

20. A readable storage medium having stored thereon one or more computer instructions, characterized in that, The instruction is executed by the processor to implement the intersection lane selection method as described in any one of claims 1 to 11.

Images