Vehicle driving track planning method, device, equipment, medium and product

Abstract

The invention discloses a vehicle driving track planning method, device and equipment, a medium and a product, and belongs to the technical field of automatic driving. The method comprises the steps of obtaining a future trajectory set of a target vehicle; according to the future trajectory set, the historical trajectories of the surrounding vehicles and the current traffic environment information, determining target predicted driving trajectories, corresponding to the driving trajectories in the future trajectory set, of the surrounding vehicles; wherein the target predicted driving track is obtained by adjusting the initial predicted driving track of the surrounding vehicles based on the drivable track; the initial predicted driving track is obtained based on a historical track and current traffic environment information; and according to the travelable trajectory and a target predicted traveling trajectory, corresponding to the travelable trajectory, of the surrounding vehicle, determining a target score of the travelable trajectory, and outputting the travelable trajectory with the highest target score in the future trajectory set. According to the invention, the autonomy and flexibility of trajectory planning of the target vehicle can be improved, and the traffic efficiency of the target vehicle can also be improved.

Description

Technical Field

[0001] The present application belongs to the field of autonomous driving technology, and in particular relates to a vehicle driving trajectory planning method, device, equipment, medium and product. Background Art

[0002] With the development of autonomous driving technology, more and more vehicles are equipped with different levels of intelligent driving functions to support assisted driving or autonomous driving.

[0003] Vehicles that support assisted or automated driving can predict the future state of their surroundings in real time, just like human drivers, to achieve safe and efficient driving. Currently, the vehicle trajectory planning systems of these vehicles are generally designed to first predict the motion trajectory of surrounding targets and then select the vehicle's own driving action. Such vehicle trajectory planning systems cause the vehicle to always be in a passive state, resulting in low traffic efficiency. For example, the vehicle's future drivable area is often occupied by surrounding traffic participants, especially in situations where there is a risk of conflict between the vehicle and other vehicles, such as in lane change scenarios, intersections, and road-to-road scenarios. After losing its drivable space, the vehicle can only slow down to avoid or wait for other vehicles to leave. It can be seen that the autonomy and flexibility of existing vehicle trajectory planning are low. Summary of the Invention

[0004] The embodiments of the present application provide a vehicle driving trajectory planning method, device, equipment, medium and product, which can improve the autonomy and flexibility of vehicle driving trajectory planning.

[0005] In a first aspect, an embodiment of the present application provides a vehicle driving trajectory planning method, the method comprising:

[0006] Get the target vehicle's future trajectory set;

[0007] When the number of drivable trajectories included in the future trajectory set is greater than one, determining target predicted driving trajectories of the surrounding vehicles corresponding to each drivable trajectory in the future trajectory set based on the future trajectory set, historical trajectories of the target vehicle's surrounding vehicles, and current traffic environment information of the target vehicle; wherein the target predicted driving trajectories of the surrounding vehicles corresponding to each drivable trajectory are obtained by adjusting initial predicted driving trajectories of the surrounding vehicles based on the drivable trajectories; and the initial predicted driving trajectories are obtained based on the historical trajectories and the current traffic environment information.

[0008] For each drivable trajectory in the future trajectory set, a target score is determined based on the drivable trajectory and the target predicted trajectories of surrounding vehicles corresponding to the drivable trajectory. The drivable trajectory with the highest target score in the future trajectory set is output.

[0009] In a second aspect, an embodiment of the present application provides a vehicle driving trajectory planning device, the device comprising:

[0010] An acquisition module is used to obtain the future trajectory set of the target vehicle;

[0011] a determination module configured to determine, when the number of drivable trajectories included in the future trajectory set is greater than one, target predicted driving trajectories of the surrounding vehicles corresponding to each drivable trajectory in the future trajectory set based on the future trajectory set, historical trajectories of the surrounding vehicles of the target vehicle, and current traffic environment information of the target vehicle; wherein the target predicted driving trajectories corresponding to each drivable trajectory of the surrounding vehicles are obtained by adjusting initial predicted driving trajectories of the surrounding vehicles based on the drivable trajectories; and the initial predicted driving trajectories are obtained based on the historical trajectories and the current traffic environment information;

[0012] The output module is configured to determine a target score for each drivable trajectory in the future trajectory set based on the drivable trajectory and the target predicted trajectories of surrounding vehicles corresponding to the drivable trajectory, and output the drivable trajectory with the highest target score in the future trajectory set.

[0013] In a third aspect, an embodiment of the present application provides a vehicle trajectory planning device, the device comprising:

[0014] a processor and a memory storing computer program instructions;

[0015] When the processor executes the computer program instructions, the vehicle driving trajectory planning method as described in the first aspect is implemented.

[0016] In a fourth aspect, an embodiment of the present application provides a computer storage medium, on which computer program instructions are stored. When the computer program instructions are executed by a processor, the vehicle driving trajectory planning method as described in the first aspect is implemented.

[0017] In a fifth aspect, an embodiment of the present application provides a computer program product. When the instructions in the computer program product are executed by a processor of an electronic device, the electronic device executes the vehicle driving trajectory planning method as described in the first aspect.

[0018] In an embodiment of the present application, a future trajectory set of the target vehicle can be obtained first. If the number of drivable trajectories included in the future trajectory set of the target vehicle is greater than 1, the target predicted driving trajectory of the surrounding vehicles corresponding to each drivable trajectory in the future trajectory set can be determined based on the future trajectory set of the target vehicle, the historical trajectories of the surrounding vehicles of the target vehicle, and the current traffic environment information of the target vehicle, wherein the target predicted driving trajectory corresponding to each drivable trajectory of the surrounding vehicles is obtained by adjusting the initial predicted driving trajectory of the surrounding vehicles based on the drivable trajectory, and the initial predicted driving trajectory of the surrounding vehicles is obtained based on the historical trajectories of the surrounding vehicles and the current traffic environment information of the target vehicle. Thereafter, based on the future trajectory set and the target predicted trajectory of the surrounding vehicles corresponding to each drivable trajectory in the future trajectory set, the target score of each drivable trajectory in the future trajectory set is obtained, and then the driving trajectory with the highest target score in the future trajectory set is output for automatic driving or assisted driving. It can be seen that the embodiment of the present application first obtains each drivable trajectories of the target vehicle; then the drivable trajectories of the target vehicle are used to predict the corresponding driving trajectories of the surrounding vehicles of the target vehicle, and the prediction of the driving trajectories of the surrounding vehicles takes into account the impact of the behavior of the target vehicle on the surrounding vehicles; finally, the drivable trajectories of the target vehicle are scored through the prediction results of the drivable trajectories of the target vehicle and the driving trajectories of its corresponding surrounding vehicles to obtain the target score of each drivable trajectory, and then the driving trajectory with the highest target score in the future trajectory set is output for automatic driving or assisted driving. In this way, the driving trajectory planning of the target vehicle can fully consider the interactivity between the target vehicle and its surrounding vehicles, which can not only improve the autonomy and flexibility of the trajectory planning of the target vehicle, but also improve the traffic efficiency of the target vehicle. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the following is a brief introduction to the drawings required for use in the embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without any creative work.

[0020] Figure 1a This is one of the schematic diagrams of the vehicle driving trajectory planning solution provided in the embodiment of the present application;

[0021] Figure 1b This is the second schematic diagram of the vehicle driving trajectory planning solution provided in the embodiment of the present application;

[0022] Figure 2 This is one of the flow charts of the vehicle driving trajectory planning method provided in the embodiment of the present application;

[0023] Figure 3 is a logical diagram of the trajectory prediction model provided in an embodiment of the present application;

[0024] Figure 4 This is the second flow chart of the vehicle driving trajectory planning method provided in the embodiment of the present application;

[0025] Figure 5 is a schematic diagram of a drivable trajectory of a target vehicle provided in an embodiment of the present application;

[0026] Figure 6 This is a schematic diagram of the structure of the vehicle driving trajectory planning device provided in an embodiment of the present application;

[0027] Figure 7 It is a structural diagram of the vehicle driving trajectory planning device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0028] The features and exemplary embodiments of various aspects of the present application will be described in detail below. In order to make the purpose, technical solutions and advantages of the present application clearer, the present application will be further described in detail below in conjunction with the accompanying drawings and specific embodiments. It should be understood that the specific embodiments described herein are only intended to explain the present application, rather than to limit the present application. For those skilled in the art, the present application can be implemented without the need for some of these specific details. The following description of the embodiments is merely to provide a better understanding of the present application by illustrating the examples of the present application.

[0029] It should be noted that, in this document, relational terms such as first and second, etc., are used only to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations. Moreover, the terms "comprises," "comprising," or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article, or device comprising a series of elements includes not only those elements, but also other elements not explicitly listed, or elements inherent to such process, method, article, or device. In the absence of further limitations, the elements defined by the phrase "comprising..." do not exclude the presence of other identical elements in the process, method, article, or device comprising the elements.

[0030] Related vehicle trajectory planning solutions such as Figure 1a As shown in the figure, the vehicle trajectory planning scheme is characterized by prediction first, then planning. Specifically, it can include the following steps: first predicting the trajectories of other vehicles (i.e., vehicles surrounding the target vehicle), then planning the behavior of the ego vehicle (i.e., the target vehicle) based on the predicted trajectories of other vehicles, and then controlling the ego vehicle based on the planned behavior.

[0031] This vehicle trajectory planning scheme lacks the interactivity between the ego vehicle's behavior and the predicted results of surrounding vehicles. This is manifested in the ego vehicle usually choosing to slow down and avoid surrounding vehicles with a certain risk of conflict, or not performing automatic lane changes when there are close vehicles to the side and rear. The vehicle trajectory planning does not take into account the actions that the ego vehicle can take and the possible reactions of surrounding vehicles under different actions. For example, in a lane change scenario, under the premise of ensuring safety, after the ego vehicle starts to change lanes, the vehicles to the side and rear will slow down to ensure that the ego vehicle can change lanes safely. Therefore, Figure 1a The vehicle trajectory planning scheme shown has low autonomy and flexibility, which can easily lead to overly conservative intelligent driving, and thus seriously affect the vehicle's traffic efficiency.

[0032] Based on this, the embodiment of the present application provides a new vehicle trajectory planning method, and its vehicle trajectory planning scheme can be but not limited to the following Figure 1b As shown, the vehicle trajectory planning scheme can be expressed as: planning first, then prediction, with the prediction process incorporating multiple planned trajectories of the ego vehicle as interactive information. Specifically, this scheme includes the following steps: first, planning the ego vehicle's behavior; then, based on its various planned behaviors, predicting the trajectories of other vehicles. Then, based on its planned behaviors and the corresponding predicted trajectories of other vehicles, the ego vehicle makes a behavioral decision. Finally, based on the ego vehicle's behavioral decision, the optimal trajectory is determined for ego vehicle control.

[0033] Compared to Figure 1a The vehicle trajectory planning scheme shown in Figure 1b The vehicle trajectory planning scheme shown can, while ensuring safety, take into account as many different behaviors as possible, such as staying straight, changing left or right lanes, and fully consider the impact of the vehicle's behavior on surrounding vehicles and the possible reactions of other traffic participants under different behaviors. This can improve the autonomy and flexibility of vehicle trajectory planning, thereby improving vehicle traffic efficiency.

[0034] The vehicle driving trajectory planning method provided by the embodiment of the present application is described in detail below through some embodiments and their application scenarios in combination with the accompanying drawings.

[0035] The vehicle driving trajectory planning method of the embodiment of the present application can be applied to a vehicle driving trajectory planning device. In specific implementation, the method can be executed by the vehicle driving trajectory planning device, or by a component of the vehicle driving trajectory planning device, such as a processor, chip, or chip system of the vehicle driving trajectory planning device, or by a logic module or software that implements all or part of the functions of the vehicle driving trajectory planning device. In actual applications, the vehicle driving trajectory planning device can be a vehicle or an electronic device, and the electronic device can be a terminal, a server, a service platform, a cloud, a distributed system, an Internet of Things, a vehicle network system, etc. Furthermore, the terminal can be a smart phone, a tablet computer, a laptop computer, a desktop computer, etc.

[0036] See also Figure 2 , Figure 2 This is one of the flow charts of the vehicle trajectory planning method provided in the embodiment of the present application. Figure 2 As shown, the vehicle driving trajectory planning method may include the following steps:

[0037] Step 201: Obtain a set of future trajectories of a target vehicle.

[0038] The future trajectory set of the target vehicle can be understood as a set of future drivable trajectories of the target vehicle, which may include one or more drivable trajectories. The specific set can be determined according to actual conditions and is not limited in this embodiment of the present application.

[0039] In an embodiment of the present application, in order to plan the driving trajectory of the target vehicle, a future trajectory set of the target vehicle may be obtained first.

[0040] However, it is worth noting that the embodiments of the present application do not limit the method for obtaining the future trajectory set of the target vehicle. In some embodiments, the future trajectory set of the target vehicle can be input by the user into the vehicle driving trajectory planning device. In other embodiments, such as Figure 1b As shown, the vehicle driving trajectory planning device can perform behavior planning on the target vehicle and obtain the future trajectory set of the target vehicle. The specific implementation thereof can be referred to the relevant content below and will not be described here.

[0041] When the number of drivable trajectories included in the future trajectory set is equal to 1, the drivable trajectory can be directly output for intelligent driving, such as assisted driving or autonomous driving.

[0042] When the number of drivable trajectories included in the future trajectory set is greater than one, step 202 may be executed to fully consider the impact of the target vehicle on surrounding vehicles and obtain prediction results of the driving trajectories of surrounding vehicles under different drivable trajectories of the target vehicle, i.e., the following target predicted driving trajectory.

[0043] The surrounding vehicles of the target vehicle (also referred to as surrounding vehicles) may include at least one of the following: the vehicle that is located in front of, behind, or closest to the target vehicle in the left adjacent lane, right adjacent lane, or current lane of the target vehicle.

[0044] Step 202: When the number of drivable trajectories included in the future trajectory set is greater than one, target predicted driving trajectories corresponding to each drivable trajectory in the future trajectory set for the surrounding vehicles are determined based on the future trajectory set, historical trajectories of the target vehicle's surrounding vehicles, and current traffic environment information of the target vehicle; wherein the target predicted driving trajectories corresponding to each drivable trajectory of the surrounding vehicles are obtained by adjusting the initial predicted driving trajectories of the surrounding vehicles based on the drivable trajectories; and the initial predicted driving trajectories are obtained based on the historical trajectories and the current traffic environment information.

[0045] In the embodiment of the present application, when predicting the driving trajectories of the target vehicle's surrounding vehicles, the impact of the target vehicle on the surrounding vehicles is fully considered. Specifically, the initial predicted driving trajectories of the surrounding vehicles can be obtained based on the historical trajectories of the surrounding vehicles and the current traffic environment information of the target vehicle. Then, the initial predicted driving trajectories of the surrounding vehicles are adjusted using each of the target vehicle's drivable trajectories to obtain target predicted driving trajectories corresponding to the surrounding vehicles and each of the drivable trajectories. In this way, the target predicted driving trajectories corresponding to the surrounding vehicles and each of the target vehicle's drivable trajectories can reflect the interactivity between the target vehicle and the surrounding vehicles, thereby improving the trajectory prediction accuracy of the target vehicle's surrounding vehicles.

[0046] For the specific acquisition of the target predicted driving trajectory corresponding to each drivable trajectory of the surrounding vehicles and the target vehicle, please refer to the following related content and will not be described here.

[0047] The embodiments of the present application do not limit the form in which the current traffic information of the target vehicle is presented. In some embodiments, the current traffic environment information of the target vehicle can be presented as (L, S, V), where L represents a set of lane line points, S represents traffic light information, and V represents road speed limit information. In other embodiments, the current traffic environment information of the target vehicle can be presented as map information of the current location of the target vehicle.

[0048] Step 203: For each drivable trajectory in the future trajectory set, determine a target score for the drivable trajectory based on the drivable trajectory and the target predicted trajectories of surrounding vehicles corresponding to the drivable trajectory, and output the drivable trajectory with the highest target score in the future trajectory set.

[0049] After obtaining each drivable trajectory of the target vehicle and the target predicted driving trajectory of the target vehicle corresponding to each driving trajectory, each drivable trajectory can be scored based on the drivable trajectory and its corresponding target predicted driving trajectory of the target vehicle. Then, the drivable trajectory with the highest score in the future trajectory set is output as the optimal driving trajectory, so that the target vehicle can perform intelligent driving based on the optimal driving trajectory, thereby improving the reliability of vehicle driving trajectory planning.

[0050] The embodiment of the present application does not limit the method for obtaining the target score of the drivable trajectory.

[0051] In some embodiments, each driving trajectory and its corresponding target predicted driving trajectory of the target vehicle can be used to perform driving simulation of the target vehicle, and the values of the target indicators of the target vehicle under different drivable trajectory simulations can be obtained. Then, based on the values of the target indicators of the target vehicle under different drivable trajectory simulations, the scores under different drivable trajectories are obtained. The target indicators can include, but are not limited to, the travel time, driving safety, and driving smoothness of the target vehicle, among which the travel time is negatively correlated with the score, and the driving safety and driving smoothness are both positively correlated with the score. That is, if the vehicle is intelligently driven based on a certain drivable trajectory, the shorter the vehicle's travel time, the higher the driving safety and driving smoothness, and the higher the score of the drivable trajectory, and vice versa.

[0052] In other embodiments, for each drivable trajectory, a preset algorithm can be used to calculate a score for the drivable trajectory based on the drivable trajectory and its corresponding target predicted driving trajectory of the target vehicle. For details, please refer to the relevant content below and will not be described here.

[0053] The vehicle driving trajectory planning method of this embodiment can first obtain a future trajectory set of the target vehicle. If the number of drivable trajectories included in the future trajectory set of the target vehicle is greater than 1, the target predicted driving trajectory of the surrounding vehicles corresponding to each drivable trajectory in the future trajectory set can be determined based on the future trajectory set of the target vehicle, the historical trajectories of the surrounding vehicles of the target vehicle, and the current traffic environment information of the target vehicle. The target predicted driving trajectory corresponding to each drivable trajectory of the surrounding vehicles is obtained by adjusting the initial predicted driving trajectory of the surrounding vehicles based on the drivable trajectory, and the initial predicted driving trajectory of the surrounding vehicles is obtained based on the historical trajectories of the surrounding vehicles and the current traffic environment information of the target vehicle. Then, based on the future trajectory set and the target predicted trajectory of the surrounding vehicles corresponding to each drivable trajectory in the future trajectory set, the target score of each drivable trajectory in the future trajectory set is obtained, and then the driving trajectory with the highest target score in the future trajectory set is output for automatic driving or assisted driving. It can be seen that the embodiment of the present application first obtains each drivable trajectories of the target vehicle; then the drivable trajectories of the target vehicle are used to predict the corresponding driving trajectories of the surrounding vehicles of the target vehicle, and the prediction of the driving trajectories of the surrounding vehicles takes into account the impact of the behavior of the target vehicle on the surrounding vehicles; finally, the drivable trajectories of the target vehicle are scored through the prediction results of the drivable trajectories of the target vehicle and the driving trajectories of its corresponding surrounding vehicles to obtain the target score of each drivable trajectory, and then the driving trajectory with the highest target score in the future trajectory set is output for automatic driving or assisted driving. In this way, the driving trajectory planning of the target vehicle can fully consider the interactivity between the target vehicle and its surrounding vehicles, which can not only improve the autonomy and flexibility of the trajectory planning of the target vehicle, but also improve the traffic efficiency of the target vehicle.

[0054] The implementation of step 202 is described in detail below.

[0055] In some embodiments, step 202 may include:

[0056] The future trajectory set, the historical trajectories of the target vehicle's surrounding vehicles, and the current traffic environment information of the target vehicle are input into the trajectory prediction model, and the target operation is performed through the trajectory prediction model to obtain the target predicted driving trajectory;

[0057] The target operations include:

[0058] Encode the future trajectory set, the historical trajectories of the target vehicle's surrounding vehicles, and the current traffic environment information of the target vehicle to obtain the target vehicle's drivable trajectory tensor corresponding to the future trajectory set, the surrounding vehicle's historical trajectory tensor corresponding to the historical trajectory, and the traffic environment tensor corresponding to the current traffic environment information;

[0059] Using the surrounding vehicle historical trajectory tensor as the first query matrix, and the traffic environment tensor as the first key matrix and the first value matrix;

[0060] generating a first predicted trajectory tensor according to the first query matrix, the first key matrix, and the first value matrix through a first cross-attention network; the first predicted trajectory tensor includes an initial predicted driving trajectory;

[0061] Using the first predicted trajectory tensor as a second query matrix, and using the target vehicle drivable trajectory tensor as a second key matrix and a second value matrix;

[0062] generating a second predicted trajectory tensor according to the second query matrix, the second key matrix, and the second value matrix through a second cross-attention network;

[0063] The second predicted trajectory tensor is decoded to obtain target predicted driving trajectories of surrounding vehicles corresponding to each drivable trajectory in the future trajectory set.

[0064] In these embodiments, the future trajectory set of the target vehicle, the historical trajectories of the target vehicle's surrounding vehicles, and the current traffic environment information of the target vehicle can be input into a trajectory prediction model (which can be simply referred to as a prediction model), and the trajectory prediction model can be used to predict the target predicted driving trajectories of the surrounding vehicles corresponding to each drivable trajectory in the future trajectory set.

[0065] The trajectory prediction model can predict the target predicted driving trajectory of the surrounding vehicles corresponding to each drivable trajectory in the future trajectory set through the target operation. The target operation may include the following steps:

[0066] Step 1: Model encoding: Encode the input target vehicle’s future trajectory set, the historical trajectory of the target vehicle’s surrounding vehicles, and the target vehicle’s current traffic environment information respectively, and obtain the target vehicle’s drivable trajectory tensor T corresponding to the future trajectory set. f , the historical trajectory tensor T of surrounding vehicles corresponding to the historical trajectory a And the traffic environment tensor T corresponding to the current traffic environment information r .

[0067] In specific implementation, the trajectory prediction model can design a first network, which converts the dimensions of the above three inputs into d m , and the dimensions are [N a ,d m ]、[N r ,d m ] and [N f ,d m ]'s 3 input tensors T a 、T rand T f , where N a Indicates the number of surrounding vehicles; N r Indicates the number of traffic information in the current traffic environment. Traffic information can be the aforementioned L, S or V, etc.; N f Indicates the number of possible trajectories in the future trajectories. The first network can be a multilayer perceptron (MLP) or a long short-term memory (LSTM) network.

[0068] Step 2: Prediction of surrounding vehicle trajectories: The trajectory prediction model can be designed with a first cross-attention network. For the surrounding vehicle historical trajectory tensor T obtained in step 1, a And the traffic environment tensor T r , the trajectory of the surrounding vehicles is predicted through the first cross attention network, and the first predicted trajectory tensor T that can represent the initial predicted driving trajectory of the surrounding vehicles is obtained m .

[0069] In specific implementation, the surrounding vehicle historical trajectory tensor T a As the first query matrix Q1, the traffic environment tensor T r The first key matrix K1 and the first value matrix V1 are calculated by the first cross attention network to obtain the first predicted trajectory tensor T m The embodiment of the present application does not limit the specific structure of the first cross attention network. In some embodiments, it can be an 8-head cross attention network, but is not limited thereto.

[0070] Step 3: Trajectory Interaction Correction: The trajectory prediction model can design a second cross-attention network to correct the target vehicle's drivable trajectory tensor T obtained in step 1. f , and the first predicted trajectory tensor T obtained in step 2 m , the predicted results can be corrected through the second cross attention network to obtain the second predicted trajectory tensor T that can characterize the target predicted driving trajectory corresponding to each drivable trajectory in the future trajectory set of surrounding vehicles e .

[0071] In specific implementation, the first predicted trajectory tensor T m As the second query matrix Q2, the target vehicle drivable trajectory tensor T f The second key matrix K2 and the second value matrix V2 are calculated by the second cross attention network to obtain the second predicted trajectory tensor T e The embodiment of the present application does not limit the specific structure of the second cross attention network. In some embodiments, it can be an 8-head cross attention network, but is not limited thereto.

[0072] Step 4: Model decoding: The trajectory prediction model can design a second network, which decodes the second predicted trajectory tensor T obtained in step 3. e Decode and obtain the target predicted driving trajectory of the surrounding vehicles corresponding to each drivable trajectory in the future trajectory set, that is, the predicted driving trajectory of the surrounding vehicles under the influence of different actions of the target vehicle The superscript f represents the drivable trajectory of the target vehicle, and the subscript t represents the future time.

[0073] For ease of understanding, the specific implementation logic of the trajectory prediction model can be found in Figure 3 .

[0074] Through the above embodiment, the trajectory prediction model can be based on the surrounding vehicle historical trajectory tensor T a As the first query matrix Q1, the traffic environment tensor T r The first key matrix K1 and the first value matrix V1 are calculated by the first cross attention network to obtain the first predicted trajectory tensor T m ; Then use the first predicted trajectory tensor T m As the second query matrix Q2, the target vehicle drivable trajectory tensor T f The second key matrix K2 and the second value matrix V2 are calculated by the second cross attention network to obtain the second predicted trajectory tensor T e , and then decode to obtain the target predicted driving trajectory of the surrounding vehicles corresponding to each drivable trajectory in the future trajectory set. In this way, by inputting the drivable trajectory of the target vehicle into the trajectory prediction model to predict the future driving trajectory of the surrounding vehicles, the impact of the target vehicle on the surrounding vehicles is fully considered, which can improve the accuracy of the driving trajectory prediction.

[0075] In some other embodiments, step 202 may include:

[0076] Determine the initial predicted driving trajectory based on historical trajectories and current traffic environment information;

[0077] For each drivable trajectory in the future trajectory set, calculating a first correlation between the initial predicted drivable trajectory and the drivable trajectory;

[0078] The initial predicted driving trajectory is adjusted according to the first correlation to obtain a target predicted driving trajectory corresponding to the drivable trajectory of the surrounding vehicles.

[0079] In these embodiments, the initial predicted driving trajectories of the surrounding vehicles may be predicted based on the historical trajectories of the surrounding vehicles and the current traffic environment information of the target vehicle.

[0080] Then, considering the interaction between the target vehicle and surrounding vehicles, the initial predicted driving trajectories of the surrounding vehicles can be adjusted using each possible trajectories in the future trajectory set to obtain the target predicted driving trajectories corresponding to each possible trajectories of the surrounding vehicles. In other words, the target predicted driving trajectory of the surrounding vehicles corresponding to a certain possible trajectories can be obtained based on the interaction between the possible trajectories and the initial predicted driving trajectories of the surrounding vehicles.

[0081] In a specific implementation, for each drivable trajectory in the future trajectory set, the correlation between the drivable trajectory and the initial predicted trajectories of the surrounding vehicles (recorded as the first correlation) can be calculated first. The correlation in the embodiment of the present application can be, but is not limited to, calculated by calculation methods such as dot product, cosine, or Euclidean.

[0082] Afterwards, the initial predicted driving trajectory is adjusted based on the first correlation to obtain the target predicted driving trajectory of the surrounding vehicles corresponding to the drivable trajectory. In the embodiment of the present application, the adjustment based on the correlation can be implemented by, but is not limited to, weighted summation or attention mechanism.

[0083] To facilitate understanding of the manner of adjusting based on correlation, the following is an example of adjusting the initial predicted driving trajectory of the target vehicle based on the first correlation.

[0084] For the weighted summation method, the first weight of the initial predicted driving trajectory can be determined based on the first correlation, and the first correlation is positively correlated with the first weight; (1-first weight) is used as the second weight of the drivable trajectory; then, the initial predicted driving trajectory and the drivable trajectory are weightedly summed using the first weight and the second weight to obtain the target predicted driving trajectory corresponding to the drivable trajectory.

[0085] For the attention mechanism, the attention weight of the initial predicted driving trajectory can be determined based on the first correlation, and the first correlation and the attention weight can be positively correlated; then, the product of the attention weight and the initial predicted driving trajectory is calculated, and then the product is added to the initial predicted driving trajectory to obtain the target predicted driving trajectory.

[0086] Through the above embodiment, after obtaining the initial predicted driving trajectory of the surrounding vehicles, for each drivable trajectory of the target vehicle, the correlation between the drivable trajectory and the initial predicted driving trajectory can be calculated, and then the initial predicted driving trajectory can be adjusted based on the correlation to obtain the target predicted driving trajectory of the surrounding vehicles corresponding to the drivable trajectory. In this way, the target predicted driving trajectory of the surrounding vehicles corresponding to the drivable trajectory can reflect the interactivity between the target vehicle and its surrounding vehicles, thereby improving the accuracy of trajectory prediction.

[0087] The embodiments of the present application do not limit the method for obtaining the initial predicted driving trajectory of the surrounding vehicles. In some embodiments, the initial predicted driving trajectory of the surrounding vehicles can be predicted by inputting the historical trajectories of the surrounding vehicles and the current traffic environment information of the target vehicle into a pre-trained network model. In other embodiments, determining the initial predicted driving trajectory based on the historical trajectories and the current traffic environment information may include:

[0088] Calculating a second correlation between the historical trajectory and the current traffic environment information;

[0089] The historical trajectory is adjusted according to the second correlation to obtain an initial predicted driving trajectory.

[0090] In these embodiments, the initial predicted driving trajectory of the target vehicle's surrounding vehicles can be obtained based on the interaction between the historical trajectories of the surrounding vehicles and the current traffic environment information of the target vehicle. In a specific implementation, the correlation between the historical trajectory and the current environment information can be first calculated (denoted as the second correlation), and then the historical trajectory can be adjusted based on the second correlation to obtain the initial predicted driving trajectory. The calculation of the second similarity and the method of adjusting the historical trajectory based on the second correlation can be referred to the aforementioned calculation of similarity and the related content of adjustment based on similarity. To avoid repetition, it will not be repeated here.

[0091] Through the above embodiment, the acquisition of the initial predicted driving trajectory of the surrounding vehicles of the target vehicle takes into account the interactivity between the surrounding vehicles and the traffic environment. In this way, the accuracy of acquiring the initial predicted driving trajectory of the surrounding vehicles of the target vehicle can be improved, thereby improving the accuracy of vehicle trajectory planning.

[0092] Regarding the aforementioned solution of obtaining a future trajectory set of the target vehicle by performing behavior planning on the target vehicle, in some embodiments, the current traffic environment information includes lane line information and traffic light information;

[0093] Get the target vehicle's future trajectory set, including:

[0094] Determine the drivable direction of the target vehicle based on lane line information and traffic light information;

[0095] According to the drivable directions, an interpolation method is used to generate a set of future trajectories.

[0096] In these embodiments, the target vehicle's possible driving directions, such as changing lanes to the left, staying straight, changing lanes to the right, etc., can be determined by first combining lane line information and traffic light information in the target vehicle's current traffic environment. Specifically, if the target vehicle's current driving lane is connected to a left adjacent lane and the traffic light includes a left turn signal, then the target vehicle's possible driving directions can be determined to include changing lanes to the left; if the target vehicle's current driving lane is connected to a right adjacent lane and the traffic light includes a right turn signal, then the target vehicle's possible driving directions can be determined to include changing lanes to the right; and if the traffic light includes a straight ahead signal, then the target vehicle's possible driving directions can be determined to include staying straight.

[0097] Afterwards, an interpolation method can be used based on the drivable directions, the current position of the target vehicle, and the lane lines corresponding to the drivable directions to generate driving trajectories corresponding to the drivable directions, thereby obtaining the drivable trajectories. The embodiments of the present application do not limit the specific form of the interpolation method. In some embodiments, the interpolation method can be a spline curve interpolation method, but is not limited to this.

[0098] In the above embodiment, the target vehicle's possible driving direction is determined by the lane line information and traffic light information in the target vehicle's current traffic environment, and then a set of future trajectories is generated. In this way, various behaviors of the vehicle can be considered as much as possible, thereby improving the success rate of target vehicle planning.

[0099] Regarding the aforementioned scheme of scoring drivable trajectories using a preset algorithm, in some embodiments, for each drivable trajectory in the future trajectory set, a target score for the drivable trajectory is determined based on the drivable trajectory and the target predicted trajectories of surrounding vehicles corresponding to the drivable trajectory, including:

[0100] Determine a safety score corresponding to the drivable trajectory based on the drivable trajectory and the target predicted trajectories of surrounding vehicles corresponding to the drivable trajectory;

[0101] Determine the speed score and stability score of the drivable trajectory according to the coordinates of each driving point in the drivable trajectory;

[0102] Determine the target score of the drivable trajectory based on the safety score, speed score, and smoothness score.

[0103] In these embodiments, based on each drivable trajectory of the target vehicle and its corresponding target predicted trajectories of surrounding vehicles, the trajectory conflict probability and / or traffic efficiency of the target vehicle and its surrounding vehicles can be predicted, and then the safety score corresponding to each drivable trajectory can be obtained, wherein the trajectory conflict probability is negatively correlated with the safety score.

[0104] In addition, based on the coordinates of each driving point in the drivable trajectory, the driving speed and driving smoothness of the drivable trajectory can be analyzed to obtain a speed score and a smoothness score, wherein the driving smoothness of the drivable trajectory is positively correlated with the smoothness score; for the speed score, a speed threshold can be set in advance, and then the speed score is determined based on the difference between the speed threshold and the driving speed of the drivable trajectory.

[0105] Afterwards, the safety score, speed score, and stability score of the drivable trajectory can be weighted averaged to obtain the target score of the drivable trajectory.

[0106] In some implementations, the target score of the drivable trajectory can be obtained by an evaluation function that considers indicators such as safety, traffic efficiency, and driving smoothness, but is not limited thereto.

[0107] In this way, the target score of the drivable trajectory can reflect information such as the safety and stability of the drivable trajectory, so that the drivable trajectory with the highest score selected based on the target score of the drivable trajectory can be the best drivable trajectory with the highest comprehensive evaluation. In this way, the reliability of the target vehicle's driving trajectory planning can be further improved.

[0108] It should be noted that the various embodiments introduced in the embodiments of the present application can be implemented in combination with each other or separately if they do not conflict with each other, and the embodiments of the present application are not limited to this.

[0109] For ease of understanding, a specific embodiment is used as an example for illustration:

[0110] In the following specific embodiment, the target vehicle is referred to as the ego vehicle, and the target predicted driving trajectory of the surrounding vehicles is realized through a model. In order to prevent the ego vehicle's drivable space from being occupied by surrounding vehicles, this specific embodiment may include the following: First, for each action that the ego vehicle may perform (such as keeping straight, changing lanes to the left, changing lanes to the right), trajectory planning is performed using a spline curve generation method. Secondly, the ego vehicle's various action trajectories are brought into the prediction model based on deep learning to calculate the prediction results of the surrounding vehicles, that is, the target predicted driving trajectory of the surrounding vehicles. Thirdly, through an evaluation function that considers indicators such as safety, traffic efficiency, and driving smoothness, combined with the prediction results of surrounding vehicles, the optimal ego vehicle behavior is selected, that is, the drivable trajectory with the highest score. Finally, the selected drivable trajectory is output to the vehicle control module to realize ego vehicle control.

[0111] Compared with related technologies, the above-mentioned specific embodiments can fully consider the impact of the ego-vehicle behavior on surrounding vehicles, improving the flexibility and efficiency of the ego-vehicle movement; can incorporate the future trajectory planned by the ego-vehicle into the prediction model to improve the accuracy of the prediction; and can calculate and process as many different ego-vehicle behaviors as possible, which can improve the success rate of ego-vehicle planning.

[0112] For easier understanding, further combined Figure 4 The implementation of the above specific embodiment is described as follows. Figure 4 As shown, the following steps may be included:

[0113] Step 1: Get the historical trajectory of surrounding vehicles in the past 2 seconds and store it in 0.1 second intervals. The historical trajectory information includes the horizontal coordinate value Vertical axis value The heading angles of surrounding vehicles t1,t2,…t end It is the timestamp and the total length of the historical track, a total of 21 time points.

[0114] Step 2: Get the current time t end The traffic environment information around the vehicle at that time (i.e., the current traffic environment information) includes the lane line point set L, traffic light information S, road speed limit information V, etc.

[0115] Step 3: Based on the traffic environment information L and S, determine the possible driving directions of the ego vehicle and use the spline curve interpolation method to generate the future 8-second trajectory in each possible driving direction (i.e., the possible driving trajectory of the target vehicle), denoted as f; Figure 5 For example, the vehicle is in a 3-lane scenario and can drive in three directions: changing lanes to the left, staying straight, and changing lanes to the right. In this case, the future trajectory set f = {f1, f2, f3}.

[0116] Step 4: Model encoding: Encode the historical trajectories of surrounding vehicles (X, Y, Ω), traffic environment information (L, S, V), and the drivable trajectory F of the vehicle. Design three MLP networks to transform the dimensions of the above input data and uniformly convert the parameter quantities of their feature dimensions into d m , and the dimensions are [N a ,d m ]、[N r ,d m ] and [N f ,d m ]'s 3 input tensors T a 、T r and T f , where N a Indicates the number of surrounding vehicles, N r Indicates the number of traffic information in the current traffic environment, N f Indicates the number of trajectories that the vehicle can travel.

[0117] Step 5: Prediction of surrounding vehicle trajectories: Using the surrounding vehicle historical trajectory tensor and traffic environment tensor encoded in Step 4, a cross attention network is designed to perform trajectory prediction calculation. In this step, the surrounding vehicle historical trajectory tensor T a As the query matrix Q1, the traffic environment tensor T r The key matrix K1 and the value matrix V1 are calculated by an 8-head cross attention network to obtain the predicted trajectory tensor T of the surrounding vehicles in the next 8 seconds. m , that is, the first predicted trajectory tensor, whose dimension is also [N a ,d m ].

[0118] Step 6: Trajectory interaction correction: Cross-attention is performed on the result tensor of Step 5 and the drivable trajectory of the vehicle, so that the target prediction structure will be affected by the different behaviors of the vehicle and the prediction results are corrected. In this step, the prediction result tensor T of Step 5 is used. m To query the matrix Q2, the vehicle's drivable trajectory tensor T f The key matrix K2 and the value matrix V2 are calculated by an 8-head cross attention network to obtain the surrounding vehicle prediction results under the influence of different actions of the vehicle, which is recorded as T e , which is the second predicted trajectory tensor.

[0119] Step 7: Model decoding: Use an MLP network to decode the result T of Step 6 e Decode and obtain the predicted output trajectory under the influence of different vehicle actions That is, the target predicted driving trajectory of the surrounding vehicles. Here, the superscript f∈{f1,f2,f3} represents the action of the ego vehicle, and the subscript t=0.1,0.2,…8.0 represents the time in the future 8 seconds, with a time interval of 0.1 seconds.

[0120] Step 8: Using a weighted evaluation function g(f), the results of Step 3 are scored based on the planned trajectory of the ego vehicle and the corresponding predicted results of surrounding vehicles, while also considering vehicle safety, ego vehicle evaluation speed, and ego vehicle driving smoothness indicators to obtain the target score for each drivable trajectory of the target vehicle.

[0121] Step 9: Output the drivable trajectory of the ego vehicle with the highest score, and end the process.

[0122] It should be noted that Figure 5 The selection of information such as time, time interval, interpolation method and attention network in the corresponding vehicle driving trajectory planning method is only an example and does not limit the form of expression of the above information. The specific selection can be made according to actual needs, and the embodiments of this application do not limit this.

[0123] Based on the vehicle driving trajectory planning method provided in the above embodiment, the present application also provides a specific implementation of a vehicle driving trajectory planning device. Please refer to the following embodiment.

[0124] See also Figure 6 The vehicle driving trajectory planning device provided in the embodiment of the present application may include:

[0125] An acquisition module 601 is used to acquire a set of future trajectories of a target vehicle;

[0126] Determination module 602 is configured to determine, when the number of drivable trajectories included in the future trajectory set is greater than one, target predicted driving trajectories of the surrounding vehicles corresponding to each drivable trajectory in the future trajectory set based on the future trajectory set, historical trajectories of the target vehicle's surrounding vehicles, and current traffic environment information of the target vehicle; wherein the target predicted driving trajectories corresponding to each drivable trajectory of the surrounding vehicles are obtained by adjusting the initial predicted driving trajectories of the surrounding vehicles based on the drivable trajectories; and the initial predicted driving trajectories are obtained based on the historical trajectories and the current traffic environment information.

[0127] Output module 603 is configured to determine a target score for each drivable trajectory in the future trajectory set based on the drivable trajectory and the target predicted trajectories of surrounding vehicles corresponding to the drivable trajectory, and output the drivable trajectory with the highest target score in the future trajectory set.

[0128] The vehicle driving trajectory planning device provided in the embodiment of the present application can implement each process in the method embodiment. To avoid repetition, it will not be described here.

[0129] Figure 7 A schematic diagram of the hardware structure of the vehicle driving trajectory planning provided by an embodiment of the present application is shown.

[0130] The vehicle driving trajectory planning device may include a processor 701 and a memory 702 storing computer program instructions.

[0131] Specifically, the processor 701 may include a central processing unit (CPU), or an application-specific integrated circuit (ASIC), or may be configured to implement one or more integrated circuits of the embodiments of the present application.

[0132] The memory 702 may include a large capacity memory for data or instructions. By way of example and not limitation, the memory 702 may include a hard disk drive (HDD), a floppy disk drive, a flash memory, an optical disk, a magneto-optical disk, a magnetic tape, or a universal serial bus (USB) drive, or a combination of two or more of these. Where appropriate, the memory 702 may include removable or non-removable (or fixed) media. Where appropriate, the memory 702 may be inside or outside the integrated gateway disaster recovery device. In a specific embodiment, the memory 702 is a non-volatile solid-state memory.

[0133] The memory may include a read-only memory (ROM), a random access memory (RAM), a magnetic disk storage medium device, an optical storage medium device, a flash memory device, an electrical, optical or other physical / tangible memory storage device. Thus, generally, the memory includes one or more tangible (non-transitory) computer-readable storage media (e.g., a memory device) encoded with software including computer-executable instructions, and when the software is executed (e.g., by one or more processors), it is operable to perform the operations described with reference to the method according to one aspect of the present disclosure.

[0134] The processor 701 implements any one of the vehicle driving trajectory planning methods in the above embodiments by reading and executing computer program instructions stored in the memory 702.

[0135] In one example, the vehicle driving trajectory planning device may further include a communication interface 707 and a bus 710. Figure 7 As shown, the processor 701, the memory 702, and the communication interface 707 are connected via a bus 710 and communicate with each other.

[0136] The communication interface 707 is mainly used to implement communication between various modules, devices, units and / or equipment in the embodiments of the present application.

[0137] The bus 710 includes hardware, software, or both, coupling the components of the vehicle trajectory planning device to each other. By way of example and not limitation,

[0138] The bus may include an Accelerated Graphics Port (AGP) or other graphics bus, an Extended Industry Standard Architecture (EISA) bus, a Front Side Bus (FSB), a Hyper Transport (HT) interconnection, an Industry Standard Architecture (ISA) bus, an InfiniBand interconnection, a Low Pin Count (LPC) bus, a memory bus, a Micro Channel Architecture (MCA) bus, a Peripheral Component Interconnect (PCI) bus, a Serial Advanced Technology Attachment (SATA) bus, a Video Electronics Standards Association Local Bus (VLB) bus or other suitable buses or a combination of two or more of these. Where appropriate, bus 710 may include one or more buses. Although the present application describes and illustrates a specific bus, the present application contemplates any suitable bus or interconnection.

[0139] In addition, in conjunction with the insulation resistance detection method in the above embodiments, embodiments of the present application may provide a computer storage medium for implementation. The computer storage medium stores computer program instructions; when the computer program instructions are executed by a processor, any one of the insulation resistance detection methods in the above embodiments is implemented.

[0140] It should be understood that the present application is not limited to the specific configurations and processes described above and illustrated in the figures. For the sake of brevity, a detailed description of known methods is omitted here. In the above embodiments, several specific steps are described and illustrated as examples. However, the method process of the present application is not limited to the specific steps described and illustrated. Those skilled in the art can make various changes, modifications, and additions, or change the order of the steps after understanding the spirit of the present application.

[0141] The functional blocks shown in the above-described block diagram can be implemented as hardware, software, firmware or a combination thereof. When implemented in hardware, it can be, for example, an electronic circuit, an application-specific integrated circuit (ASIC), appropriate firmware, a plug-in, a function card, etc. When implemented in software, the elements of the present application are programs or code segments that are used to perform the required tasks. The program or code segment can be stored in a machine-readable medium, or transmitted on a transmission medium or a communication link by a data signal carried in a carrier wave. "Machine-readable medium" can include any medium that can store or transmit information. Examples of machine-readable media include electronic circuits, semiconductor memory devices, ROM, flash memory, erasable ROM, floppy disks, CD-ROMs, optical disks, hard disks, optical fiber media, radio frequency links, etc. The code segment can be downloaded via a computer network such as the Internet, an intranet, etc.

[0142] It should also be noted that the exemplary embodiments mentioned in this application describe some methods or systems based on a series of steps or devices. However, this application is not limited to the order of the above steps. In other words, the steps can be performed in the order mentioned in the embodiments, or in a different order, or several steps can be performed simultaneously.

[0143] Aspects of the present disclosure have been described above with reference to the flowcharts and / or block diagrams of the methods, devices (systems) and computer program products according to the embodiments of the present disclosure. It should be understood that each box in the flowchart and / or block diagram and the combination of each box in the flowchart and / or block diagram can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer or other programmable data processing device to produce a machine so that these instructions executed by the processor of the computer or other programmable data processing device enable the implementation of the function / action specified in one or more boxes of the flowchart and / or block diagram. Such a processor can be, but is not limited to, a general-purpose processor, a special-purpose processor, a special application processor or a field programmable logic circuit. It is also understood that each box in the block diagram and / or flowchart and the combination of the boxes in the block diagram and / or flowchart can also be implemented by dedicated hardware that performs the specified function or action, or can be implemented by a combination of dedicated hardware and computer instructions.

[0144] The above description is only a specific embodiment of the present application. Those skilled in the art will clearly understand that for the convenience and brevity of description, the specific working processes of the systems, modules and units described above can refer to the corresponding processes in the aforementioned method embodiments, and will not be repeated here. It should be understood that the scope of protection of the present application is not limited thereto. Any person skilled in the art can easily think of various equivalent modifications or replacements within the technical scope disclosed in the present application, and these modifications or replacements should be included in the scope of protection of the present application.

Claims

1. A vehicle driving trajectory planning method, characterized in that: include: Get the target vehicle's future trajectory set; When the number of drivable trajectories included in the future trajectory set is greater than one, determining a target predicted driving trajectory of the surrounding vehicle corresponding to each drivable trajectory in the future trajectory set based on the future trajectory set, historical trajectories of vehicles surrounding the target vehicle, and current traffic environment information of the target vehicle; wherein the target predicted driving trajectory corresponding to each drivable trajectory of the surrounding vehicle is obtained by adjusting an initial predicted driving trajectory of the surrounding vehicle based on the drivable trajectory; and the initial predicted driving trajectory is obtained based on the historical trajectories and the current traffic environment information; For each drivable trajectory in the future trajectory set, a target score of the drivable trajectory is determined based on the drivable trajectory and the target predicted trajectories of the surrounding vehicles corresponding to the drivable trajectory, and the drivable trajectory with the highest target score in the future trajectory set is output.

2. The method according to claim 1, characterized in that The step of determining, based on the future trajectory set, historical trajectories of surrounding vehicles of the target vehicle, and current traffic environment information of the target vehicle, a target predicted driving trajectory of the surrounding vehicles corresponding to each drivable trajectory in the future trajectory set includes: Inputting the future trajectory set, the historical trajectories of the target vehicle's surrounding vehicles, and the current traffic environment information of the target vehicle into a trajectory prediction model, performing a target operation through the trajectory prediction model, and obtaining the target predicted driving trajectory; The target operation includes: Encoding the future trajectory set, the historical trajectories of the target vehicle's surrounding vehicles, and the current traffic environment information of the target vehicle to obtain a drivable trajectory tensor of the target vehicle corresponding to the future trajectory set, a historical trajectory tensor of the surrounding vehicles corresponding to the historical trajectory, and a traffic environment tensor corresponding to the current traffic environment information; Using the surrounding vehicle historical trajectory tensor as a first query matrix, and using the traffic environment tensor as a first key matrix and a first value matrix; generating, by a first cross-attention network, a first predicted trajectory tensor according to the first query matrix, the first key matrix, and the first value matrix; the first predicted trajectory tensor including the initial predicted driving trajectory; Using the first predicted trajectory tensor as a second query matrix, and using the target vehicle drivable trajectory tensor as a second key matrix and a second value matrix; generating, by a second cross-attention network, a second predicted trajectory tensor according to the second query matrix, the second key matrix, and the second value matrix; The second predicted trajectory tensor is decoded to obtain target predicted driving trajectories of the surrounding vehicles corresponding to each drivable trajectory in the future trajectory set.

3. The method according to claim 1, characterized in that The step of determining, based on the future trajectory set, historical trajectories of surrounding vehicles of the target vehicle, and current traffic environment information of the target vehicle, a target predicted driving trajectory of the surrounding vehicles corresponding to each drivable trajectory in the future trajectory set includes: Determining the initial predicted driving trajectory based on the historical trajectory and the current traffic environment information; For each drivable trajectory in the future trajectory set, calculating a first correlation between the initial predicted driving trajectory and the drivable trajectory; The initial predicted driving trajectory is adjusted according to the first correlation to obtain a target predicted driving trajectory of the surrounding vehicles corresponding to the drivable trajectory.

4. The method according to claim 3, characterized in that The determining the initial predicted driving trajectory according to the historical trajectory and the current traffic environment information includes: Calculating a second correlation between the historical trajectory and the current traffic environment information; The historical trajectory is adjusted according to the second correlation to obtain the initial predicted driving trajectory.

5. The method according to claim 1, wherein The current traffic environment information includes lane line information and traffic light information; The obtaining of the future trajectory set of the target vehicle includes: Determining a drivable direction of the target vehicle based on the lane line information and the traffic light information; The future trajectory set is generated using an interpolation method according to the drivable directions.

6. The method according to claim 1, characterized in that For each drivable trajectory in the future trajectory set, determining a score of the drivable trajectory according to the drivable trajectory and the target predicted trajectories of the surrounding vehicles corresponding to the drivable trajectory includes: Determining a safety score corresponding to the drivable trajectory based on the drivable trajectory and the target predicted trajectories of the surrounding vehicles corresponding to the drivable trajectory; determining a speed score and a stability score of the drivable trajectory according to the coordinates of each driving point in the drivable trajectory; A target score for the drivable trajectory is determined based on the safety score, the vehicle speed score, and the smoothness score.

7. A vehicle trajectory planning device, characterized in that: The device comprises: An acquisition module is used to obtain the future trajectory set of the target vehicle; a determination module configured to, when the number of drivable trajectories included in the future trajectory set is greater than one, determine, based on the future trajectory set, historical trajectories of vehicles surrounding the target vehicle, and current traffic environment information of the target vehicle, a target predicted driving trajectory of the surrounding vehicle corresponding to each drivable trajectory in the future trajectory set; wherein the target predicted driving trajectory of the surrounding vehicle corresponding to each drivable trajectory is obtained by adjusting an initial predicted driving trajectory of the surrounding vehicle based on the drivable trajectory; and the initial predicted driving trajectory is obtained based on the historical trajectories and the current traffic environment information; an output module, configured to determine, for each drivable trajectory in the future trajectory set, a target score for the drivable trajectory based on the drivable trajectory and the target predicted trajectories of the surrounding vehicles corresponding to the drivable trajectory, and output the drivable trajectory in the future trajectory set with the highest target score.

8. A vehicle trajectory planning device, characterized in that: The device includes: a processor and a memory storing computer program instructions; when the processor executes the computer program instructions, the vehicle driving trajectory planning method according to any one of claims 1 to 6 is implemented.

9. A computer-readable storage medium, characterized in that The computer-readable storage medium stores computer program instructions, and when the computer program instructions are executed by a processor, the vehicle driving trajectory planning method according to any one of claims 1 to 6 is implemented.

10. A computer program product, characterized in that When the instructions in the computer program product are executed by a processor of an electronic device, the electronic device executes the vehicle driving trajectory planning method as described in any one of claims 1 to 6.

Images