An automatic driving vehicle path planning method, device, equipment and medium

Abstract

The application discloses an automatic driving vehicle path planning method, device, equipment and medium. The method comprises the following steps: determining prior features of at least one obstacle based on sensing information of an unmanned vehicle; determining weights of at least two candidate regions matched based on a position of the vehicle according to a motion state of the vehicle, and determining a sorting result of the obstacles according to the weights of the candidate regions, the prior features of the obstacles and positions of the obstacles; determining a preset number of dangerous obstacles among the obstacles according to the sorting result, and taking the prior features of the dangerous obstacles as inputs of a prediction model to determine a planned path of the automatic driving vehicle. The technical scheme solves the problems of low accuracy and high time cost caused by that the prediction model directly extracts abstract features from the sensing information, and can improve the prediction accuracy and model generation efficiency of the prediction model and enhance the explainability of the prediction model.

Description

Technical Field

[0001] This invention relates to the field of autonomous driving technology, and in particular to a method, apparatus, device, and medium for autonomous vehicle path planning. Background Technology

[0002] With the continuous advancement of technology, the need for navigation capabilities for autonomous vehicles in various road conditions is becoming increasingly urgent. The navigation systems of autonomous vehicles need to acquire perception information such as radar and vision, and make reasonable decisions based on the recognition results of this perception information, so as to reach the destination in a timely and accurate manner while ensuring driving safety.

[0003] Currently, navigation systems for autonomous vehicles mainly rely on prediction models such as Lyft and Waymo, which directly represent perceived information as features using deep learning networks and make driving decisions based on the abstract features extracted from the perceived information.

[0004] However, the existing technology of directly representing perceptual information through deep learning networks requires a lot of time to train the deep learning networks, and the stability of the training effect is difficult to guarantee, resulting in poor interpretability of the prediction model. Summary of the Invention

[0005] This invention provides a method, apparatus, device, and medium for path planning of autonomous vehicles, which solves the problems of low accuracy and high time cost caused by the prediction model directly extracting abstract features from perceived information. It can improve the prediction accuracy and model generation efficiency of the prediction model while enhancing the interpretability of the prediction model.

[0006] According to one aspect of the present invention, an autonomous vehicle path planning method is provided, the method comprising:

[0007] Based on the vehicle's perception information, determine at least one prior feature of an obstacle;

[0008] Based on the vehicle's motion state, determine the weights of matching at least two candidate regions divided based on the vehicle's position, and determine the ranking of obstacles based on the weights of each candidate region, the prior features of each obstacle, and the position of each obstacle.

[0009] Based on the sorting results, a preset number of dangerous obstacles are identified among the obstacles, and the prior features of each dangerous obstacle are used as input to the prediction model to determine the planned path of the autonomous vehicle.

[0010] According to another aspect of the present invention, an autonomous vehicle path planning device is provided, the device comprising:

[0011] The prior feature determination module is used to determine the prior features of at least one obstacle based on the vehicle's perception information;

[0012] The sorting result determination module is used to determine the matching weights of at least two candidate regions divided based on the vehicle's position according to the vehicle's motion state, and to determine the sorting result of the obstacles according to the weights of each candidate region, the prior features of each obstacle, and the position of each obstacle.

[0013] The path planning and determination module is used to determine a preset number of dangerous obstacles among the obstacles based on the sorting results, and to use the prior features of each dangerous obstacle as input to the prediction model to determine the planned path of the autonomous vehicle.

[0014] According to another aspect of the present invention, an electronic device is provided, the electronic device comprising:

[0015] At least one processor; and

[0016] A memory communicatively connected to the at least one processor; wherein,

[0017] The memory stores a computer program that can be executed by the at least one processor, which enables the at least one processor to perform the autonomous vehicle path planning method according to any embodiment of the present invention.

[0018] According to another aspect of the present invention, a computer-readable storage medium is provided, the computer-readable storage medium storing computer instructions for causing a processor to execute and implement the autonomous vehicle path planning method according to any embodiment of the present invention.

[0019] The technical solution of this invention determines the prior features of at least one obstacle using the vehicle's perception information; then, based on the vehicle's motion state, it determines the matching weights of at least two candidate regions divided based on the vehicle's position, and determines the ranking of obstacles based on the weights of each candidate region, the prior features of each obstacle, and the position of each obstacle; furthermore, based on the ranking of obstacles, it identifies a preset number of dangerous obstacles among them, and uses the prior features of each dangerous obstacle as input to a prediction model to determine the planned path of the autonomous vehicle. This solution solves the problems of low accuracy and high time cost caused by the prediction model directly abstracting features from perception information, and can improve the prediction accuracy and model generation efficiency while enhancing the interpretability of the prediction model.

[0020] It should be understood that the description in this section is not intended to identify key or essential features of the embodiments of the present invention, nor is it intended to limit the scope of the invention. Other features of the invention will become readily apparent from the following description. Attached Figure Description

[0021] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0022] Figure 1 This is a flowchart of an autonomous vehicle path planning method provided in Embodiment 1 of the present invention;

[0023] Figure 2 This is a schematic diagram of the structure of a prediction model provided according to an embodiment of the present invention;

[0024] Figure 3 This is a flowchart of an autonomous vehicle path planning method provided in Embodiment 2 of the present invention;

[0025] Figure 4 This is a schematic diagram of candidate region division in a simple driving scenario provided by an embodiment of the present invention;

[0026] Figure 5 This is a schematic diagram of candidate region division in a complex driving scenario provided by an embodiment of the present invention;

[0027] Figure 6 This is a schematic diagram of the structure of an autonomous vehicle path planning device according to Embodiment 3 of the present invention;

[0028] Figure 7 This is a schematic diagram of the structure of an electronic device that implements the autonomous vehicle path planning method of this invention. Detailed Implementation

[0029] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present invention.

[0030] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this invention are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be used interchangeably where appropriate so that the embodiments of the invention described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, apparatus, product, or device that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or devices. The acquisition, storage, use, and processing of data in the technical solutions of this application all comply with the relevant provisions of national laws and regulations.

[0031] Example 1

[0032] Figure 1 This is a flowchart illustrating a path planning method for autonomous vehicles according to Embodiment 1 of the present invention. This embodiment is applicable to path planning scenarios for autonomous vehicles, and is particularly suitable for complex traffic scenarios such as congestion and multi-lane traffic. This method can be executed by an autonomous vehicle path planning device, which can be implemented in hardware and / or software and can be configured in an electronic device. Figure 1 As shown, the method includes:

[0033] S110. Determine at least one prior feature of an obstacle based on the vehicle's perception information.

[0034] This solution can be executed by the controller of an autonomous vehicle. The controller can acquire multi-dimensional perception information through sensors such as radar and cameras deployed on the vehicle. This perception information can include the vehicle's speed, the speed of obstacles, the distance between the vehicle and obstacles, and response time. The controller can identify the type of obstacle based on driving scene images captured by the cameras, such as static obstacles and moving obstacles. Static obstacles can include lane lines, green belts, and buildings, while dynamic obstacles can include vehicles, pedestrians, and bicycles. Based on the vehicle's perception information, the controller can determine the vehicle's motion characteristics, navigation map characteristics, and obstacle characteristics. Obstacle characteristics include prior features of the obstacles, such as the obstacle's safety loss and the rate of change of safety loss. Specifically, the controller can determine the safety loss of each obstacle based on the vehicle's speed, the speed of obstacles, the distance between the vehicle and obstacles, and response time. Based on the vehicle's speed and the distance between the vehicle and obstacles, the controller can determine the rate of change of safety loss of each obstacle.

[0035] S120. Based on the motion state of the vehicle, determine the weights of matching at least two candidate regions divided based on the vehicle's position, and determine the ranking result of the obstacles based on the weights of each candidate region, the prior features of each obstacle, and the position of each obstacle.

[0036] Understandably, autonomous vehicles can divide their perception range into regions based on their current location, obtaining at least two candidate regions. The controller can further divide the vehicle's perception range into candidate regions based on the complexity of the driving scenario. After determining each candidate region based on the vehicle's current location, the controller can determine the matching weights for each candidate region based on the vehicle's motion state. The motion state of an autonomous vehicle can include driving straight, making a U-turn, turning left, turning right, and reversing. The controller can set one or more sets of candidate region weights for each motion state, with each set including the weights of each candidate region. Based on the location of each obstacle, the controller can determine the candidate region to which each obstacle belongs. Based on the weights of the candidate regions to which each obstacle belongs and the prior features of each obstacle, the controller can calculate the attention level of each obstacle and rank them to obtain the ranking results. It is easy to understand that the prior features safety loss and the rate of change of safety loss can characterize the degree of danger of an obstacle. The degree of danger of candidate regions associated with the vehicle's motion state will increase due to the vehicle's motion. Therefore, based on the prior characteristics of each obstacle and the weight of the candidate region to which each obstacle belongs, the danger of each obstacle can be assessed in real time and accurately.

[0037] S130. Based on the sorting results, a preset number of dangerous obstacles are identified among the obstacles, and the prior features of each dangerous obstacle are used as input to the prediction model to determine the planned path of the autonomous vehicle.

[0038] The controller of an autonomous vehicle can select a preset number of obstacles with high risk assessment values ​​as dangerous obstacles based on the ranking results. For example, the top 64 moving targets with the highest risk assessment values ​​in the ranking results can be identified as dangerous obstacles. The controller can also select obstacles with risk assessment values ​​greater than a preset assessment threshold as dangerous obstacles based on the ranking results. For example, obstacles with risk assessment values ​​greater than 0.8 in the ranking results can be identified as dangerous obstacles.

[0039] The controller can use the vehicle's motion characteristics, navigation map features, and features of various hazards as inputs to the prediction model to obtain prediction information for the autonomous vehicle. The vehicle's motion characteristics can include its position, direction of travel, and speed. Navigation map features can include road features of the driving environment, such as road boundary lines and reference lines. Hazard features can include prior features of hazards, as well as their position, length, width, direction of travel, and distance from the vehicle.

[0040] In this solution, the prediction model can be built based on a deep learning algorithm, and the prediction information can include the vehicle's speed, acceleration, curvature, and position. Based on the prediction information, the controller can generate a planned path for the autonomous vehicle to ensure safe driving.

[0041] Specifically, the prediction model can be pre-trained using training sample data, which can include multiple sets to ensure good training results. Each set of training sample data can include feature data and label data. Similar to the application process of the prediction model, the feature data can include vehicle motion features, navigation map features, and a preset number of dangerous obstacle features. The label data can be vehicle motion information matched with the feature data.

[0042] Figure 2 This is a schematic diagram of the structure of a prediction model provided according to an embodiment of the present invention. Figure 2 As shown, the prediction model may include an input branching structure, a multi-head self-attention structure, a multilayer perceptron structure, and a long short-term memory structure. The input branching structure can be used to extract features from vehicle motion characteristics, hazard obstacle characteristics, and navigation map features. The multi-head self-attention structure can be used to extract relationship features between various input features, thereby improving the accuracy and efficiency of the prediction model. The multilayer perceptron structure can compress features. The long short-term memory structure can extract temporal features between multiple frames of input data.

[0043] In a preferred embodiment, the Long Short-Term Memory (LSTM) structure can use the current frame and four historical frames as input data to extract temporal features between the input data, thereby achieving good prediction results.

[0044] In a feasible scheme, the input branch structure includes a first input branch, a second input branch, and a third input branch; the first input branch, the second input branch, and the third input branch are respectively used to extract features from the vehicle's motion features, dangerous obstacle features, and navigation map features; the input branch structure, the multi-head self-attention structure, the multilayer perceptron structure, and the long short-term memory structure are sequentially connected.

[0045] As is easily understood, each input branch in the input branch structure is used to extract features from the vehicle's motion features, hazard obstacle features, and navigation map features, respectively. Each input branch can include at least one fully connected layer, and the structures of the input branches can be the same or different. The first input branch, the second input branch, and the third input branch can extract features from the vehicle's motion features, obstacle features, and navigation map features, respectively, to obtain the first feature extraction result, the second feature extraction result, and the third feature extraction result.

[0046] The controller can input the first, second, and third feature extraction results into a multi-head self-attention structure according to a preset input method to extract the relationship features between the input features. After obtaining the output features with relationship features from the multi-head self-attention structure, the controller can use a multilayer perceptron mechanism to further extract deeper features from the output features, and input the extracted deep features into a long short-term memory structure to extract the temporal features between deep features across multiple frames.

[0047] The prediction model structure of this scheme can realize multi-dimensional feature extraction of vehicle motion characteristics, dangerous obstacle characteristics, and navigation map characteristics, which is conducive to improving the robustness and accuracy of the model.

[0048] Based on the above scheme, optionally, the multi-head self-attention structure may include three inputs; wherein the first input and the second input are both determined based on the first feature extraction result of the first input branch, the second feature extraction result of the second input branch, and the third feature extraction result of the third input branch; the third input is determined based on the first feature extraction result of the first input branch.

[0049] Specifically, the controller can fuse the first feature extraction result, the second feature extraction result, and the third feature extraction result according to a preset fusion method, and use the fused feature as the first input or the second input. The first input and the second input can be the same or different. The controller can also directly concatenate the first feature extraction result, the second feature extraction result, and the third feature extraction result in a certain order to obtain the fused feature. For example, using A, B, and C to represent the first feature extraction result, the second feature extraction result, and the third feature extraction result respectively, the fused feature can be represented as [A, B, C].

[0050] The controller can also sequentially select partial feature extraction results from the first feature extraction result, the second feature extraction result, and the third feature extraction result, and alternately concatenate different feature extraction results to generate a fused feature. For example, the fused feature can be represented as [A1, B1, C1, A2, B2, C2], where A1 represents the first part of the feature extraction result in the first feature extraction result, A2 represents the second part of the feature extraction result in the first feature extraction result, and B1, C1, B2, and C2 are similarly represented.

[0051] Understandably, the controller can also use the first feature extraction result as a third input to determine, based on the first, second, and third inputs, the output feature that relates the vehicle's motion features, navigation map features, and hazardous obstacle features. The calculation formula for the output feature can be expressed as:

[0052]

[0053] Where K represents the first input, V represents the second input, Q represents the third input, and d k represents the dimension of the first input, softmax represents the normalized exponential function, and Attention(Q,K,V) represents the output features of the multi-head attention mechanism.

[0054] Prediction models with the aforementioned multi-head self-attention structure can improve prediction efficiency and accuracy.

[0055] This solution does not directly input the vehicle's perception information into the prediction model to extract abstract features of obstacles for driving decision-making. Instead, it determines the prior features of obstacles based on the perception information, uses these prior features as the basis for driving decisions, and inputs them into the prediction model to obtain the vehicle's driving decision. Therefore, this solution can improve the accuracy and efficiency of the prediction model's decision-making. This technical solution determines the prior features of at least one obstacle using the vehicle's perception information; then, based on the vehicle's motion state, it determines the matching weights of at least two candidate regions divided based on the vehicle's position, and determines the obstacle ranking based on the weights of each candidate region, the prior features of each obstacle, and the position of each obstacle; then, based on the obstacle ranking, it identifies a preset number of dangerous obstacles among the obstacles, and uses the prior features of each dangerous obstacle as input to the prediction model to determine the planned path for the autonomous vehicle. This solution solves the problems of low accuracy and high time cost caused by the prediction model directly extracting abstract features from the perception information, and can improve the prediction accuracy and model generation efficiency while enhancing the interpretability of the prediction model.

[0056] Example 2

[0057] Figure 3This is a flowchart of an autonomous vehicle path planning method provided in Embodiment 2 of the present invention. Based on the above embodiments, this embodiment refines the calculation of the safety factor of dynamic obstacles and the selection of dangerous obstacles. For example... Figure 3 As shown, the method includes:

[0058] S210. Based on the vehicle's perception information, determine the safety factor between the vehicle and at least one obstacle; wherein the safety factor is the ratio of the current distance between the vehicle and the obstacle to the safe distance.

[0059] Based on the vehicle's perception information, the controller can calculate the safety factor between the vehicle and each obstacle within the perception range. This safety factor can be used to assess the degree of danger posed by the obstacle to the vehicle. Specifically, the safety factor can be represented as the ratio of the current distance between the vehicle and the obstacle to a safe distance, which can be calculated based on the vehicle's speed, the obstacle's speed, and the vehicle's response time to driving decisions.

[0060] In this solution, the controller can determine the safe distance between the vehicle and each obstacle based on the vehicle's speed information, the obstacle's speed information, and the response time; and determine the safety factor between the vehicle and each obstacle based on the safe distance and the distance information between the vehicle and the obstacle. The safety factor may include a longitudinal safety factor and a lateral safety factor.

[0061] Understandably, given the varying directions of motion of dynamic obstacles in driving scenarios, the controller can analyze the relationship between the vehicle and each obstacle from both lateral and longitudinal perspectives. Based on information such as vehicle speed, vehicle acceleration, obstacle speed, obstacle acceleration, and response time, the controller can calculate the lateral safety distance between the vehicle and each obstacle. Based on this lateral safety distance, the controller can calculate the lateral safety factor for each obstacle. Similarly, based on the same information, the controller can also calculate the longitudinal safety distance between the vehicle and each obstacle. Based on this longitudinal safety distance, the controller can calculate the longitudinal safety factor for each obstacle.

[0062] 1. The process of determining the longitudinal safety distance may include:

[0063] (1) Determine the relative motion direction between the vehicle and the dynamic obstacle based on the vehicle's speed information and the speed information of the dynamic obstacle.

[0064] Intuitively, the vehicle's speed information indicates its direction of movement, and the speed information of a dynamic obstacle indicates its direction of movement. For longitudinal safety distance, the controller can determine the relative direction of movement between the vehicle and the dynamic obstacle based on their respective directions of travel. Based on the vehicle's lateral and longitudinal directions, the speed of the dynamic obstacle is decomposed, and the relative direction of movement between the vehicle and the obstacle is determined based on the decomposed speed of the obstacle in the vehicle's longitudinal direction. The controller can set different longitudinal safety distance calculation methods based on the relative direction of movement between the vehicle and the dynamic obstacle.

[0065] This solution determines the relative motion direction between the vehicle and dynamic obstacles by using the motion direction indicated by speed information. Then, based on different relative motion directions, it calculates the longitudinal safety distance between the vehicle and each dynamic obstacle using the vehicle's speed information, the speed information of the dynamic obstacles, and the response time. This solution achieves accurate longitudinal safety distance calculation, which helps ensure the accuracy of safety factor calculation.

[0066] (2) Determine the longitudinal safe distance between the vehicle and each dynamic obstacle based on the relative motion direction, the vehicle's speed information, the speed information of the dynamic obstacle, and the response time.

[0067] It is understood that the relative motion direction can include both the same direction and opposite directions; the velocity information can include velocity, lateral velocity information and longitudinal velocity information; wherein, the longitudinal velocity information can include longitudinal velocity, longitudinal deceleration and longitudinal acceleration; the lateral velocity information can include lateral velocity.

[0068] <1> If the relative motion direction between the vehicle and the dynamic obstacle is the same, the longitudinal safe distance between the vehicle and the dynamic obstacle is determined based on the longitudinal velocity of the preceding traffic target, the longitudinal velocity of the following traffic target, the maximum longitudinal deceleration, the maximum longitudinal acceleration, the minimum longitudinal deceleration, and the response time; wherein the maximum longitudinal deceleration, the maximum longitudinal acceleration, and the minimum longitudinal deceleration are determined based on the statistical results of the traffic target attribute parameters.

[0069] If the relative motion direction between the vehicle and the dynamic obstacle is the same, the controller can calculate the longitudinal safety distance between the vehicle and the dynamic obstacle based on the longitudinal velocity of the preceding traffic target, the longitudinal velocity of the following traffic target, the maximum longitudinal deceleration, the maximum longitudinal acceleration, the minimum longitudinal deceleration, and the response time. The preceding and following traffic targets are different; the preceding traffic target could be the vehicle itself or a dynamic obstacle, and similarly, the following traffic target could be either the vehicle or a dynamic obstacle. The maximum longitudinal deceleration, maximum longitudinal acceleration, and minimum longitudinal deceleration are determined based on statistical results of traffic target attribute parameters. These traffic target attribute parameters can include parameters such as the traffic target's category, model, size, load range, acceleration range, and deceleration range. The controller can acquire the attribute parameters of traffic targets existing in the traffic environment, perform parameter statistics for each type of traffic target, and obtain the maximum longitudinal deceleration, maximum longitudinal acceleration, and minimum longitudinal deceleration for each type of traffic target. The controller can determine the maximum longitudinal deceleration, maximum longitudinal acceleration, and minimum longitudinal deceleration of a dynamic obstacle based on the maximum longitudinal deceleration, maximum longitudinal acceleration, and minimum longitudinal deceleration corresponding to the traffic target of the category to which the dynamic obstacle belongs.

[0070] In a specific scheme, the controller can determine the maximum displacement of the following traffic target within the response time based on its longitudinal velocity, maximum longitudinal acceleration, and response time. Based on the longitudinal velocity of the following traffic target, the longitudinal velocity, maximum longitudinal acceleration, minimum longitudinal deceleration, maximum longitudinal deceleration, and response time of the preceding traffic target, the controller can determine the maximum braking distance between the preceding and following traffic targets within the response time. The maximum displacement of the following traffic target and the maximum braking distance are then added together to obtain the safety distance. Specifically, when the relative motion direction between the vehicle and the dynamic obstacle is the same, the formula for calculating the longitudinal safety distance can be expressed as:

[0071]

[0072] Among them, v r v represents the longitudinal velocity of the following traffic target. f The longitudinal velocity of the preceding traffic target is represented by t, and the response time is represented by a. lon,max,accel a represents the maximum longitudinal acceleration. lon,min,brake a represents the minimum longitudinal deceleration. lon,max,brake This represents the maximum longitudinal deceleration.

[0073] <2> If the relative motion directions of the vehicle and the dynamic obstacle are opposite, the longitudinal safe distance between the vehicle and the dynamic obstacle is determined based on the longitudinal speed of the vehicle, the longitudinal speed of the dynamic obstacle, the minimum longitudinal deceleration, the maximum longitudinal acceleration, the minimum longitudinal deceleration of the vehicle, and the response time.

[0074] If the relative motion directions of the vehicle and the dynamic obstacle are opposite, the controller can determine the longitudinal safe distance between the vehicle and the dynamic obstacle based on the vehicle's longitudinal speed, the dynamic obstacle's longitudinal speed, minimum longitudinal deceleration, maximum longitudinal acceleration, the vehicle's minimum longitudinal deceleration, and the response time.

[0075] Taking a vehicle traveling in the forward direction and a dynamic obstacle traveling in the reverse direction as an example, the controller can determine the vehicle's maximum speed within the response time based on the vehicle's longitudinal velocity, maximum longitudinal acceleration, and response time. Similarly, the controller can determine the dynamic obstacle's maximum speed within the response time based on the obstacle's longitudinal velocity, maximum longitudinal acceleration, and response time. Based on the vehicle's longitudinal velocity, maximum speed within the response time, minimum longitudinal deceleration, and response time, the controller can calculate the vehicle's maximum displacement. Based on the dynamic obstacle's longitudinal velocity, maximum speed within the response time, minimum longitudinal deceleration, and response time, the controller can calculate the dynamic obstacle's maximum displacement. The maximum displacement of the vehicle and the maximum displacement of the dynamic obstacle are added together to obtain the safety distance. Similarly, when the vehicle is traveling in the reverse direction and the dynamic obstacle is traveling in the forward direction, and the relative motion directions of the vehicle and the dynamic obstacle are opposite, the formula for calculating the longitudinal safety distance can be expressed as:

[0076]

[0077] v 1lon,t =v 1lon +ta lon,max,accel ;

[0078] v 2lon,t =|v 2lon |+ta lon,max,accel ;

[0079] Among them, v 1lon v represents the speed of a traffic target traveling in the forward direction. 2lon Let t represent the speed of the traffic target traveling in the opposite direction, and a represent the response time. lon,max,accel a represents the maximum longitudinal acceleration. lon,min,brake a represents the minimum longitudinal deceleration. lon,min,brake,correct This indicates the minimum longitudinal deceleration of the vehicle.

[0080] 2. The process of determining the lateral safety distance may include:

[0081] (1) Determine the safe lateral distance between the vehicle and each dynamic obstacle based on the vehicle's lateral velocity, the lateral velocity of the dynamic obstacle, the maximum lateral acceleration, the minimum lateral deceleration, and the response time.

[0082] Understandably, the lateral velocity information also includes lateral acceleration and lateral deceleration. For lateral safety distance, the controller can determine the lateral safety distance between the vehicle and each dynamic obstacle based on the relative position of the vehicle and the obstacle. The controller can use the lateral displacement of two traffic targets approaching each other with maximum lateral acceleration and braking with minimum lateral deceleration until they stop as the lateral safety distance.

[0083] In a specific scheme, the formula for calculating the lateral safety distance can be expressed as:

[0084]

[0085] v 1lat,t =v 1lat +ta lat,max,accel ;

[0086] v 2lat,t =v 2lat -ta lat,max,accel ;

[0087] Where μ represents the fault tolerance space, v 1lat v represents the lateral velocity of the traffic target located on the left. 2lat The lateral velocity of the traffic target on the right is represented by t, and the response time is represented by a. lat,max,accel a represents the maximum lateral acceleration. lat,min,brake This represents the minimum lateral deceleration.

[0088] This scheme calculates the safety distance from both lateral and longitudinal perspectives, which facilitates accurate calculation of the safety factor and maximizes the driving safety of the vehicle. In some implementations, μ is 0.2m, adding a margin of error to the calculated safety distance to ensure driving safety.

[0089] After obtaining the lateral and longitudinal safe distances between the vehicle and the dynamic obstacle, the lateral and longitudinal safety factors can be calculated using the following formulas:

[0090]

[0091]

[0092] Where lon_dis represents the current longitudinal distance between the vehicle and the dynamic obstacle, d minThe longitudinal safety distance is represented by `lon_safe_coeff`, which represents the longitudinal safety coefficient; `lat_dis` represents the current lateral distance between the vehicle and a dynamic obstacle, and `d` represents the lateral distance between the vehicle and the obstacle. min,lat lat_safe_coeff represents the lateral safety distance.

[0093] In this scheme, optionally, the prior features also include the rate of change of safety loss; wherein the rate of change of safety loss is determined based on the distance and relative speed between the vehicle and the dynamic obstacle.

[0094] The rate of change of safety loss can include the rate of change of lateral safety loss and the rate of change of longitudinal safety loss. The rate of change of lateral safety loss can be determined based on the ratio of the current lateral relative speed of the vehicle to the current lateral distance between the vehicle and the dynamic obstacle, and the rate of change of longitudinal safety loss can be determined based on the ratio of the current longitudinal relative speed of the vehicle to the current longitudinal distance between the vehicle and the dynamic obstacle.

[0095] This scheme can calculate the rate of change of safety loss and incorporate it into the prior characteristics of dynamic obstacles, which is beneficial for assessing the hazard of dynamic obstacles from two dimensions: safety loss and the change of safety loss.

[0096] S220. Based on each safety factor and the inverted Gaussian model, determine the safety loss that matches each obstacle.

[0097] Understandably, the controller of an autonomous vehicle can use an inverted Gaussian model to constrain each safety factor to the range of (0,1) to obtain the normalized safety loss for each obstacle matching.

[0098] Specifically, if the safety factor of an obstacle is less than a preset threshold, the safety loss matching the obstacle is determined based on an inverted Gaussian model. According to the calculation method of the safety factor in S210, the greater the current distance relative to the safe distance, the greater the safety factor, and the lower the danger level of the obstacle. If the safety factor is greater than or equal to the preset threshold, it means that for this vehicle, the obstacle corresponding to the safety factor is not dangerous or has a low degree of danger. The controller can directly set the safety loss of obstacles with no danger or low danger to 0, without needing to calculate the safety loss.

[0099] If the safety factor is less than the preset threshold, the controller can constrain the safety factor of each obstacle to the range (0,1) according to the inverted Gaussian model, so as to accelerate the convergence speed of the prediction model during training and improve the driving decision accuracy of the prediction model.

[0100] The controller can determine the longitudinal safety loss based on the longitudinal safety factor and the lateral safety loss based on the lateral safety factor, based on the inverted Gaussian model; and determine the safety loss matching the obstacle based on the longitudinal safety loss, the lateral safety loss and the predetermined weight coefficients; wherein the weight coefficients can be determined based on the rate of change of safety loss.

[0101] Specifically, the formulas for calculating longitudinal safety loss and lateral safety loss can be expressed as follows:

[0102] lon_safe_cost=-1×[1-gaussian(K1-lon_safe_coeff)];

[0103] lat_safe_cost=-1×[1-gaussian(K2-lat_safe_coeff)];

[0104] Where lon_safe_cost represents the longitudinal safety loss, lon_safe_coeff represents the longitudinal safety coefficient, lat_safe_cost represents the lateral safety loss, lat_safe_coeff represents the lateral safety coefficient, gaussian represents the Gaussian distribution, and K1 and K2 represent preset coefficient thresholds, which can be the same or different.

[0105] The controller can determine the overall safety loss of the operational target based on the lateral safety loss, longitudinal safety loss, and weighting coefficients. Specifically, the formula for calculating the safety loss can be expressed as:

[0106] safe_cost=w_lon_safe×lon_safe_cost+w_lat_safe×lat_safe_cost;

[0107] Where w_lon_safe represents the weighting coefficient of longitudinal safety loss, and w_lat_safe represents the weighting coefficient of lateral safety loss.

[0108] The weighting coefficients can be obtained statistically from historical safety loss data. For example, the weighting coefficient for lateral safety loss can be set to 0.54, and the weighting coefficient for longitudinal safety loss can be set to 0.46. Alternatively, the weighting coefficients can be determined based on the safety loss change rate. The controller can determine the safety loss change rate based on the current distance between the vehicle and the obstacle, as well as the current speed. The safety loss change rate can include both lateral and longitudinal safety loss change rates. Specifically, the lateral safety loss change rate can be determined based on the ratio of the current lateral speed to the current lateral distance between the vehicle and the obstacle, and the longitudinal safety loss change rate can be determined based on the ratio of the current longitudinal speed to the current longitudinal distance between the vehicle and the obstacle.

[0109] The formulas for calculating the longitudinal safety loss change rate and the lateral safety loss change rate can be expressed as follows:

[0110]

[0111]

[0112] Where lon_speed represents the relative speed between the vehicle and the dynamic obstacle in the longitudinal direction, and lat_speed represents the relative speed between the vehicle and the dynamic obstacle in the lateral direction.

[0113] This scheme allows for the setting of matching weighting coefficients for lateral and longitudinal safety losses to calculate the overall safety loss of obstacles, which is beneficial for achieving accurate assessment of obstacle safety.

[0114] S230. Based on the vehicle's location, determine at least two candidate areas.

[0115] After obtaining the prior features of each obstacle, the controller can divide the vehicle's perception range into candidate regions based on the vehicle's position. The region division method may vary depending on the driving scenario. Figure 4 This is a schematic diagram of candidate region division in a simple driving scenario provided by an embodiment of the present invention. Figure 4 As shown, in a narrow one-way driving scenario, the controller can divide the vehicle's current position into three candidate regions: candidate region ①, candidate region ②, and candidate region ③. The controller can also divide the vehicle's perception range into five candidate regions based on distance, for example, candidate region ①, candidate region ②, candidate region ③, candidate region ④, and candidate region ⑤.

[0116] Figure 5 This is a schematic diagram of candidate region division in a complex driving scenario according to an embodiment of the present invention. In complex driving scenarios such as traffic congestion, the vehicle control device can divide the candidate region ⑤ where the vehicle is located into regions such as... Figure 5 The nine candidate regions shown can be configured by the vehicle control equipment to set the size of each candidate region according to the driving scenario.

[0117] It should be noted that the above-mentioned candidate region division based on the complexity of the driving scenario is only one method of candidate region division. The controller can also divide candidate regions based on factors such as the lane the vehicle is in and the road segment the vehicle is traveling on. This embodiment does not limit the method of candidate region division. The area of ​​each candidate region can be the same or different, and the shape of each candidate region can be a regular shape or an irregular shape. This embodiment does not limit the number, size, or shape of candidate regions.

[0118] S240. Based on the motion state of the vehicle, determine the weight of each candidate region according to the preset weight allocation principle.

[0119] The motion states of an autonomous vehicle can include going straight, making a U-turn, turning left, turning right, and reversing. The controller can set the weights of each candidate region for motion state matching based on the degree of influence of each motion state on each candidate region. For example... Figure 5 Taking the candidate region division method shown as an example, when the vehicle is traveling straight, the affected candidate regions can include ③, ⑤, ⑥, and ⑨. The degree of influence of candidate regions ③, ⑤, ⑥, and ⑨ can be ordered as ⑤, ⑥, ③, and ⑨. Therefore, the weights of candidate regions ①-⑨ matching the straight-traveling state can be 1, 1, 1.5, 1, 2, 1.8, 1, 1, and 1.5, respectively. The controller can also gradually increase the weight of the target candidate region based on the order in which each candidate region is affected by the vehicle's motion state. (Continuing with the example...) Figure 5 Taking the candidate region division method shown as an example, when the vehicle is in a left-turn state, it will affect candidate regions ⑤, ⑥, ③, ② and ① in sequence. Then the weights of candidate regions ①-⑨ matched in the left-turn state can be 1.2, 1.4, 1.6, 1, 2, 1.8, 1, 1 and 1 respectively.

[0120] S250. Determine the candidate area to which each obstacle belongs based on its location.

[0121] The controller can compare the location of each obstacle with the range of each candidate area to determine the candidate area to which each obstacle belongs.

[0122] S260. Determine the ranking of obstacles based on the weight of each candidate region, the candidate region to which each obstacle belongs, and the safety loss of each obstacle.

[0123] In a feasible solution, determining the obstacle ranking result based on the weight of each candidate region, the candidate region to which each obstacle belongs, and the safety loss of each obstacle includes:

[0124] Determine the weight of the candidate region to which each obstacle belongs, and determine the weighted safety loss based on the weight of the candidate region to which each obstacle belongs and the safety loss of each obstacle;

[0125] The ranking of obstacles is determined based on the weighted safety loss of each obstacle.

[0126] The controller can determine the weight of the safety loss matching for each obstacle based on the weight of each candidate region and the candidate region to which each obstacle belongs. By multiplying the safety loss of each obstacle by its matching weight, the controller can obtain the weighted safety loss for each obstacle. The controller can then sort the weighted safety losses of each obstacle and output the sorted results.

[0127] This scheme can determine the weighted safety loss of each obstacle, which is beneficial to the reliability of obstacle hazard assessment.

[0128] S270. Based on the sorting results, select a preset number of obstacles as dangerous obstacles in descending order of weighted safety loss.

[0129] The controller can determine the degree of danger of each obstacle based on the weighted safety loss in the obstacle sorting results, and select a preset number of obstacles with relatively high degree of danger as dangerous obstacles.

[0130] The above scheme selects a certain number of obstacles as dangerous obstacles, which helps to improve the safety and reliability of driving decisions while ensuring the timeliness of driving decisions.

[0131] S280. Use the prior features of each dangerous obstacle as input to the prediction model to determine the planned path of the autonomous vehicle.

[0132] This technical solution determines the prior features of at least one obstacle using the vehicle's perception information. Then, based on the vehicle's motion state, it determines the matching weights of at least two candidate regions divided based on the vehicle's position. Based on the weights of each candidate region, the prior features of each obstacle, and the position of each obstacle, it determines the obstacle ranking. Next, based on the obstacle ranking, it identifies a preset number of dangerous obstacles and uses the prior features of each dangerous obstacle as input to the prediction model to determine the planned path for the autonomous vehicle. This solution solves the problems of low accuracy and high time cost caused by the prediction model directly abstracting features from perception information. It can improve the prediction accuracy and model generation efficiency of the prediction model while enhancing its interpretability.

[0133] Example 3

[0134] Figure 6This is a schematic diagram of the structure of an autonomous vehicle path planning device provided in Embodiment 3 of the present invention. Figure 6 As shown, the device includes:

[0135] The prior feature determination module 310 is used to determine the prior features of at least one obstacle based on the perception information of the vehicle.

[0136] The sorting result determination module 320 is used to determine the matching weights of at least two candidate regions divided based on the position of the vehicle according to the motion state of the vehicle, and to determine the sorting result of the obstacles according to the weight of each candidate region, the prior features of each obstacle and the position of each obstacle;

[0137] The path planning and determination module 330 is used to determine a preset number of dangerous obstacles among the obstacles based on the sorting results, and to use the prior features of each dangerous obstacle as input to the prediction model to determine the planned path of the autonomous vehicle.

[0138] In this scheme, optionally, the prior features include security losses;

[0139] The prior feature determination module 310 includes:

[0140] A safety factor determination unit is used to determine the safety factor between the vehicle and at least one obstacle based on the vehicle's perception information; wherein the safety factor is the ratio of the current distance between the vehicle and the obstacle to a safe distance;

[0141] The safety loss determination unit is used to determine the safety loss matching each obstacle based on each safety factor and an inverted Gaussian model.

[0142] In one feasible solution, the sorting result determination module 320 includes:

[0143] The candidate region determination unit is used to determine at least two candidate regions based on the position of the vehicle.

[0144] The weight determination unit is used to determine the weight of each candidate region according to the motion state of the vehicle and the preset weight allocation principle.

[0145] Based on the above scheme, optionally, the sorting result determination module 320 includes:

[0146] The region determination unit is used to determine the candidate region to which each obstacle belongs based on its location.

[0147] The sorting result determination unit is used to determine the sorting result of obstacles based on the weight of each candidate region, the candidate region to which each obstacle belongs, and the safety loss of each obstacle.

[0148] In this embodiment, optionally, the sorting result determination unit includes:

[0149] The weighted loss determination subunit is used to determine the weight of the candidate region to which each obstacle belongs, and to determine the weighted safety loss based on the weight of the candidate region to which each obstacle belongs and the safety loss of each obstacle.

[0150] The sorting result determination sub-unit is used to determine the sorting result of obstacles based on the weighted safety loss of each obstacle.

[0151] In a preferred embodiment, the path planning and determination module 330 includes:

[0152] A hazardous obstacle determination unit is used to determine a preset number of obstacles as hazardous obstacles from among the obstacles based on the sorting results.

[0153] The path planning and determination unit is used to take the prior features of each dangerous obstacle as input to the prediction model in order to determine the planned path of the autonomous vehicle.

[0154] Based on the above scheme, optionally, the prior feature also includes the rate of change of safety loss; wherein the rate of change of safety loss is determined based on the distance and relative speed between the vehicle and the obstacle.

[0155] The autonomous vehicle path planning device provided in this embodiment of the invention can execute the autonomous vehicle path planning method provided in any embodiment of the invention, and has the corresponding functional modules and beneficial effects of the method.

[0156] Example 4

[0157] Figure 7 A schematic diagram of an electronic device 410 that can be used to implement embodiments of the present invention is shown. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as personal digital processors, cellular phones, smartphones, wearable devices (e.g., helmets, glasses, watches, etc.), and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely illustrative and are not intended to limit the implementation of the invention described and / or claimed herein.

[0158] like Figure 7As shown, the electronic device 410 includes at least one processor 411 and a memory, such as a read-only memory (ROM) 412 or a random access memory (RAM) 413, communicatively connected to the at least one processor 411. The memory stores computer programs executable by the at least one processor. The processor 411 can perform various appropriate actions and processes based on the computer program stored in the ROM 412 or loaded from storage unit 418 into the RAM 413. The RAM 413 may also store various programs and data required for the operation of the electronic device 410. The processor 411, ROM 412, and RAM 413 are interconnected via a bus 414. An input / output (I / O) interface 415 is also connected to the bus 414.

[0159] Multiple components in electronic device 410 are connected to I / O interface 415, including: input unit 416, such as keyboard, mouse, etc.; output unit 417, such as various types of displays, speakers, etc.; storage unit 418, such as disk, optical disk, etc.; and communication unit 419, such as network card, modem, wireless transceiver, etc. Communication unit 419 allows electronic device 410 to exchange information / data with other devices through computer networks such as the Internet and / or various telecommunications networks.

[0160] Processor 411 can be a variety of general-purpose and / or special-purpose processing components with processing and computing capabilities. Some examples of processor 411 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various special-purpose artificial intelligence (AI) computing chips, various processors running machine learning model algorithms, a digital signal processor (DSP), and any suitable processor, controller, microcontroller, etc. Processor 411 performs the various methods and processes described above, such as path planning methods for autonomous vehicles.

[0161] In some embodiments, the autonomous vehicle path planning method may be implemented as a computer program tangibly contained in a computer-readable storage medium, such as storage unit 418. In some embodiments, part or all of the computer program may be loaded into and / or installed on electronic device 410 via ROM 412 and / or communication unit 419. When the computer program is loaded into RAM 413 and executed by processor 411, one or more steps of the autonomous vehicle path planning method described above may be performed. Alternatively, in other embodiments, processor 411 may be configured to perform the autonomous vehicle path planning method by any other suitable means (e.g., by means of firmware).

[0162] Various embodiments of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field-programmable gate arrays (FPGAs), application-specific integrated circuits (ASICs), application-specific standard products (ASSPs), systems-on-a-chip (SoCs), payload-programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various embodiments may include implementations in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which may be a dedicated or general-purpose programmable processor, capable of receiving data and instructions from a storage system, at least one input device, and at least one output device, and transmitting data and instructions to the storage system, the at least one input device, and the at least one output device.

[0163] Computer programs used to implement the methods of the present invention may be written in any combination of one or more programming languages. These computer programs may be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device, such that when executed by the processor, the computer programs cause the functions / operations specified in the flowcharts and / or block diagrams to be performed. The computer programs may be executed entirely on a machine, partially on a machine, or as a standalone software package, partially on a machine and partially on a remote machine, or entirely on a remote machine or server.

[0164] In the context of this invention, a computer-readable storage medium can be a tangible medium that may contain or store a computer program for use by or in conjunction with an instruction execution system, apparatus, or device. A computer-readable storage medium may include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination thereof. Alternatively, a computer-readable storage medium may be a machine-readable signal medium. More specific examples of machine-readable storage media include electrical connections based on one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fibers, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof.

[0165] To provide interaction with a user, the systems and techniques described herein can be implemented on an electronic device having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and pointing device (e.g., a mouse or trackball) through which the user provides input to the electronic device. Other types of devices can also be used to provide interaction with the user; for example, feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including sound input, voice input, or tactile input).

[0166] The systems and technologies described herein can be implemented in computing systems that include backend components (e.g., as data servers), or computing systems that include middleware components (e.g., application servers), or computing systems that include frontend components (e.g., user computers with graphical user interfaces or web browsers through which users can interact with implementations of the systems and technologies described herein), or any combination of such backend, middleware, or frontend components. The components of the system can be interconnected via digital data communication of any form or medium (e.g., communication networks). Examples of communication networks include local area networks (LANs), wide area networks (WANs), blockchain networks, and the Internet.

[0167] A computing system can include clients and servers. Clients and servers are generally located far apart and typically interact through communication networks. The client-server relationship is created by computer programs running on the respective computers and having a client-server relationship with each other. The server can be a cloud server, also known as a cloud computing server or cloud host, which is a hosting product within the cloud computing service system to address the shortcomings of traditional physical hosts and VPS services, such as high management difficulty and weak business scalability.

[0168] It should be understood that the various forms of processes shown above can be used, with steps reordered, added, or deleted. For example, the steps described in this invention can be executed in parallel, sequentially, or in different orders, as long as the desired result of the technical solution of this invention can be achieved, and this is not limited herein.

[0169] The specific embodiments described above do not constitute a limitation on the scope of protection of this invention. Those skilled in the art should understand that various modifications, combinations, sub-combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this invention should be included within the scope of protection of this invention.

Claims

1. A path planning method for autonomous vehicles, characterized in that, The method includes: Based on the vehicle's perception information, at least one prior feature of an obstacle is determined; the prior feature includes a safety loss; the safety loss is determined based on the ratio of the current distance between the vehicle and the obstacle to a safe distance. Based on the vehicle's location, at least two candidate areas must be identified; The weight of each candidate region is determined according to the weight allocation principle of the vehicle's motion state matching; the motion state includes going straight, making a U-turn, turning left, turning right, and reversing; Based on the location of each obstacle, determine the candidate region to which each obstacle belongs; Determine the weight of the candidate region to which each obstacle belongs, and determine the weighted safety loss based on the weight of the candidate region to which each obstacle belongs and the safety loss of each obstacle; The ranking of obstacles is determined based on the weighted safety loss of each obstacle; Based on the sorting results, a preset number of dangerous obstacles are identified among the obstacles, and the prior features of each dangerous obstacle are used as input to the prediction model to determine the planned path of the autonomous vehicle.

2. The method according to claim 1, characterized in that, The determination of prior features of at least one obstacle based on the vehicle's perception information includes: Based on the vehicle's perception information, determine the safety factor between the vehicle and at least one obstacle; wherein the safety factor is the ratio of the current distance between the vehicle and the obstacle to the safe distance; Based on each safety factor, and using the inverted Gaussian model, the safety loss matching each obstacle is determined.

3. The method according to claim 1, characterized in that, The step of determining a preset number of dangerous obstacles from among the obstacles based on the sorting results, and using the prior features of each dangerous obstacle as input to the prediction model to determine the planned path of the autonomous vehicle, includes: Based on the sorting results, a preset number of obstacles are selected as dangerous obstacles in descending order of weighted safety loss; The prior features of each dangerous obstacle are used as input to the prediction model to determine the planned path of the autonomous vehicle.

4. The method according to claim 1, characterized in that, The prior features also include the rate of change of safety loss; wherein the rate of change of safety loss is determined based on the distance and relative speed between the vehicle and the obstacle.

5. A path planning device for an autonomous vehicle, characterized in that, The device includes: A priori feature determination module is used to determine the priori features of at least one obstacle based on the vehicle's perception information; the priori features include safety loss; the safety loss is determined based on the ratio of the current distance between the vehicle and the obstacle to a safe distance; The sorting result determination module is used to determine at least two candidate regions based on the vehicle's position; determine the weight of each candidate region according to the weight allocation principle matching the vehicle's motion state; the motion state includes going straight, making a U-turn, turning left, turning right, and reversing; determine the candidate region to which each obstacle belongs based on the position of each obstacle; determine the weight of each obstacle's candidate region, and determine the weighted safety loss based on the weight of each obstacle's candidate region and the safety loss of each obstacle; and determine the sorting result of the obstacles based on the weighted safety loss of each obstacle. The path planning and determination module is used to determine a preset number of dangerous obstacles among the obstacles based on the sorting results, and to use the prior features of each dangerous obstacle as input to the prediction model to determine the planned path of the autonomous vehicle.

6. An electronic device, characterized in that, The electronic device includes: At least one processor; and A memory communicatively connected to the at least one processor; wherein, The memory stores a computer program that can be executed by the at least one processor, the computer program being executed by the at least one processor to enable the at least one processor to perform the autonomous vehicle path planning method according to any one of claims 1-4.

7. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer instructions that cause a processor to execute the autonomous vehicle path planning method according to any one of claims 1-4.

Images