An automatic driving cooperative motion planning method fusing interaction trajectory prediction

Abstract

The application relates to a cooperative motion planning method fusing interactive trajectory prediction, and belongs to the field of automatic driving vehicles.The method comprises the following steps: S1, constructing a polynomial parameterized trajectory generation model meeting vehicle kinematic constraints, generating a future driving action set based on environment information and a current position state of the vehicle; S2, constructing an interactive prediction-decision planning model based on deep learning, extracting interactive hidden features based on bidirectional interactive behaviors of an automatic driving vehicle and surrounding vehicles, and predicting surrounding vehicle trajectories; and S3, constructing an adaptive trajectory generation model, building a learnable trajectory selection module, scoring the driving action set by using the trajectory prediction result of step S2, combining a bias generation network to optimize the result, and finally generating an optimal trajectory.The application can interactively model future states of the ego vehicle and surrounding intelligent agents, predict future trajectories of the surrounding vehicles under different planning interactions, and thus improve prediction accuracy.

Description

Technical Field

[0001] The present invention belongs to the field of autonomous driving vehicles and relates to an autonomous driving collaborative motion planning method integrating interactive trajectory prediction. Background Art

[0002] Thanks to advances in computer vision, deep learning, sensor fusion, and motion planning, autonomous driving technology has rapidly developed in recent years. By autonomously operating vehicles, this technology is gradually transitioning from assisted driving with limited human intervention to fully autonomous driving without human intervention. Autonomous driving systems generally include four major modules: environmental perception and fusion, obstacle trajectory prediction, joint decision-making and planning, and trajectory tracking and control. These modules work together to achieve the driving task. Joint decision-making and planning plays a key role in ensuring safety and is the core embodiment of the system's intelligence.

[0003] Currently, autonomous vehicle technology is primarily categorized into three types: modular autonomous driving, modular end-to-end autonomous driving, and fully end-to-end autonomous driving. Modular autonomous driving is a traditional and relatively mature autonomous driving technology. This technology decomposes autonomous driving tasks into multiple independent functional modules, including perception, prediction, decision-making and planning, and control, enabling each module to independently perform its corresponding functions based on different task requirements. Its advantage lies in the clear display of the output of each module, facilitating flexible adjustment and optimization. However, due to the relative independence of each sub-module, modular autonomous driving can result in data transmission delays and accumulated errors, which can pose safety risks. Therefore, it is typically applied to Level 2 assisted driving systems. However, in complex traffic environments, the two-way interaction between the ego vehicle and surrounding agents (such as other vehicles and pedestrians) further complicates traffic situations. The ego vehicle's future trajectory is not only influenced by surrounding vehicles, but its driving intentions can also alter the future states of surrounding vehicles. These factors increase the risk of prediction, decision-making, and planning failures in modular autonomous driving due to accumulated errors and interactive behaviors, potentially leading to safety accidents.

[0004] Therefore, there is an urgent need for a safe motion planning method that can effectively deal with the interaction between vehicles to ensure that the autonomous driving system is safer and more reliable in dynamic and complex environments. Summary of the Invention

[0005] In light of this, the present invention aims to provide a collaborative motion planning method for autonomous driving that integrates interactive trajectory prediction. This method interactively models the future states of the ego vehicle and surrounding intelligent agents, predicting the future trajectories of surrounding vehicles under different planning interactions, thereby improving prediction accuracy. Simultaneously, the ego vehicle can mimic the behavioral decisions of human drivers, resulting in safer, more efficient, more comfortable, and more human-like motion planning trajectories. The entire process of this method is differentiable.

[0006] A collaborative motion planning method for autonomous driving integrated with interactive trajectory prediction includes the following steps:

[0007] S1: Construct a polynomial parameterized trajectory generation model that satisfies the vehicle's kinematic constraints and generates a set of future driving actions based on environmental information and the vehicle's current position.

[0008] S2: Build an interactive prediction-decision-making planning model based on deep learning. Based on the two-way interaction between the autonomous vehicle and surrounding vehicles, extract the hidden features of the interaction and predict the trajectories of surrounding vehicles.

[0009] S3: Build an adaptive trajectory generation model and a learnable trajectory selection module. Use the trajectory prediction results of step S2 to score the driving action set, and combine the deviation to generate network optimization results to finally generate the optimal trajectory.

[0010] Furthermore, in step S1, a polynomial parameterized trajectory generation model is constructed, which specifically includes the following steps:

[0011] S11: Obtain local information of the vehicle: According to the current position of the vehicle and the high-precision map information provided, obtain the current motion state P of the vehicle c And the future drivable lane path point R, namely:

[0012] P c =[x c ,y c ,θ c ,v c ,a c ]

[0013] R=[L1,L2,...,L n ]

[0014]

[0015] Among them, x c ,y c ,θ c ,v c ,a c They represent the horizontal coordinate, vertical coordinate, heading angle, speed and acceleration of the current position of the vehicle, respectively.n and They represent the nth future drivable lane and the hth path point information of the lane, The horizontal coordinate, vertical coordinate, navigation angle, curvature and speed limit of the h-th path point in the n-th lane are represented respectively;

[0016] S12: Constructing a polynomial parameterized trajectory generation model: Based on the vehicle's current position information and future drivable path information provided in step S11, and in order to adapt to the complex structure of urban road scenes, a trajectory model of lateral displacement and longitudinal displacement versus time is constructed using a quintic polynomial and a quartic polynomial in the Frenet coordinate system, respectively:

[0017] Δs(t)=b0+b1t+b2t 2 +b3t 3 +b4t 4

[0018] Δd(t)=r0+r1t+r2t 2 +r3t 3 +r4t 4 +r5t 5

[0019] Among them, Δd(t) and Δs(t) represent the lateral and longitudinal displacements at time t, respectively. [b0, b1, ..., b4] and [r0, r1, ..., r4] are the parameters of the polynomial parameterized model, which can be obtained by the current position information P of the vehicle. c and the terminal position information P e Solve, P e The calculation can be performed by assuming different driving speeds; finally, the obtained Δd(t) and Δs(t) are converted to the Cartesian coordinate system to obtain the final planned trajectory set

[0020] Furthermore, in step S2, an interactive prediction-decision planning model based on deep learning is constructed, which specifically includes the following steps:

[0021] S21: Interactive prediction-decision planning model input: The input of the interactive prediction-decision planning model mainly includes the historical trajectory characteristics of surrounding vehicles The local map information where the vehicle is currently located And the vehicle planning trajectory set obtained in step S12

[0022]

[0023] Among them, X i and They are respectively represented as the feature information and position features of the i-th intelligent agent, namely the vehicle, th Represents the historical time domain, l i ,w i Respectively represent the horizontal coordinate, vertical coordinate, heading angle, lateral speed, longitudinal speed, vehicle length and width at time t, M n Represents the nth local lane information, They are represented as the horizontal coordinate, vertical coordinate, navigation angle, lane type, speed limit and traffic light status from the first waypoint to the jth waypoint respectively;

[0024] S22: Input feature encoding: Encode the agent features obtained in step S21 using the LSTM network to obtain hidden features Map feature information is encoded using the MLP network (multi-layer perceptron network) to obtain hidden features The planned trajectory set is encoded using another MLP network to obtain hidden features

[0025] S23: Global feature encoding: In order to obtain the global feature information of the agent and the local map, the encoding results of the agent and map obtained in step S22 are encoded with global features using the self-attention mechanism of the two-layer Transformer;

[0026] G featrue =Attention(Q F ,K F ,V F ,mask)

[0027]

[0028] G feature ={G agent ,G map}

[0029] Among them, G featrue is a global feature, which can be decomposed into the agent global feature G agent and the global map feature G map , Q F ,K F ,V F ,mask are the query,key, value and corresponding mask of the attention mechanism, respectively. Attention and concatenate represent the attention mechanism and concatenation, respectively;

[0030] S24: Interactive feature information modeling: Combine the hidden features of the planning trajectory set obtained in step S22 The global feature G of the agent obtained in step S23 agent and the global map feature G map, use the cross attention mechanism to model the interaction of each predicted agent i:

[0031]

[0032]

[0033] Among them, F i,am and F i,ae They are respectively represented as the interaction features between the ith agent and the map and the planned trajectory, Q i,F , mask m , mask e Represent the query, key, value and corresponding mask required to calculate attention, respectively. Represents the global characteristics of the i-th agent;

[0034] S25: Prediction trajectory decoding: Combine the agent-map interaction feature F obtained in step S24 i,am and the interactive features F of the agent-planning trajectory i,ae , using an MLP network to decode the future trajectory of the i-th agent

[0035]

[0036] Among them, t f Indicates the future time domain.

[0037] Furthermore, in step S3, an adaptive trajectory generation model is constructed, which specifically includes the following steps:

[0038] S31: Learnable cost weight: Combined with the hidden features of the vehicle's historical trajectory obtained in step S22 and the vehicle-map interactive hidden feature F obtained in step S24 e,am Splice and use an MLP network to decode the weight of the cost function

[0039]

[0040] Among them, τ s ,τ e ,τ c are expressed as the weights of safety, efficiency and comfort respectively;

[0041] S32: Trajectory scoring modeling: Combine the weights obtained in step S31 Construct trajectory scoring model;

[0042] U o =τ s Us +τ e U e +τ c U c

[0043]

[0044] Among them, U o represents the trajectory score, U s ,U e ,U c denote the cost functions of safety, efficiency and comfort respectively, o is the index of the predicted agent in the safety cost function, S safety is the safety threshold, They represent the horizontal and vertical coordinates of the planned trajectory of the vehicle and the predicted trajectory of the agent o in the future, They represent the speed, lateral and longitudinal acceleration of the vehicle’s planned trajectory in the future, v limit Indicates lane speed limit;

[0045] S33: Trajectory deviation optimization network: Combine the planned trajectory scores obtained in step S32 and select the trajectory with the highest score As the basic trajectory of the trajectory deviation optimization network, find the corresponding hidden features of the ego-vehicle planning trajectory Combined with the hidden features of the vehicle's historical trajectory obtained in step S22 and the vehicle-map interactive hidden feature F obtained in step S24 e,am , use the MLP network to decode the coordinate deviation of the trajectory and add it to the trajectory to generate the optimal trajectory;

[0046] S34: Loss function formulation: In order to make trajectory prediction more accurate and the driving behavior of the vehicle more human-like, a loss function is designed

[0047]

[0048] in, They represent the prediction trajectory imitation loss, planning trajectory imitation loss and planning selection score loss respectively, λ1, λ2, λ3 represent the weights corresponding to each loss, G i ,G e Represent the true trajectories of the predicted agent i and the ego vehicle, represents the planning trajectory obtained by the model, smoothhl1 and softmax are smooth L1 loss and activation functions respectively, y z is the planning trajectory score, a and Z are the number of predicted agents and the number of planned trajectories respectively.

[0049] The beneficial effects of the present invention are:

[0050] 1) This paper designs a polynomial parameterized model. Based on the current position of the vehicle and the characteristics of the local environment, a polynomial curve is used to generate a set of planned trajectories that can be exercised in the future. In combination with the kinematic model, a candidate trajectory is generated that is more consistent with the actual driving conditions.

[0051] 2) This paper designs an interactive prediction-decision-making planning model based on deep learning. This model can extract the historical trajectory characteristics of the intelligent agent, the environmental characteristics and the planned trajectory characteristics of the vehicle, use the Transformer network to establish the interactive characteristics of the vehicle-map and vehicle-planning trajectory, and generate a predicted trajectory under interactive behavior.

[0052] 3) This paper designs an adaptive trajectory generation model that uses the historical characteristics of the ego vehicle and its interaction characteristics with the map to decode the adaptive trajectory selection weights. It then selects the adaptive trajectory based on safety, efficiency, and comfort scores, and combines it with a bias network model to generate the optimal trajectory. Compared to traditional modular autonomous driving, this model considers the impact of two-way interactions between the ego vehicle and surrounding vehicles, as well as the cumulative error between models, to ensure safe driving.

[0053] Other advantages, objects, and features of the present invention will be described in part in the following description and, in part, will be apparent to those skilled in the art upon examination of the following description or may be learned from practice of the present invention. The objects and other advantages of the present invention may be realized and obtained through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0054] In order to make the purpose, technical solutions and advantages of the present invention more clear, the present invention will be described in detail below with reference to the accompanying drawings, in which:

[0055] Figure 1 This is a logical framework diagram for the overall implementation of the deep learning-based interactive motion planning method for autonomous driving of the present invention. DETAILED DESCRIPTION

[0056] The following describes the embodiments of the present invention by means of specific examples, and those skilled in the art can easily understand other advantages and effects of the present invention from the contents disclosed in this specification. The present invention can also be implemented or applied through other different specific embodiments, and the details in this specification can also be modified or changed in various ways based on different viewpoints and applications without departing from the spirit of the present invention. It should be noted that the illustrations provided in the following embodiments are only schematic illustrations of the basic concept of the present invention, and the following embodiments and features in the embodiments can be combined with each other without conflict.

[0057] See also Figure 1The present invention designs a collaborative motion planning method for autonomous driving that integrates interactive trajectory prediction. Taking into account the problem of two-way vehicle interaction in complex traffic environments, a polynomial parameterized model is constructed to generate a set of planned trajectories. Combining the current map information of the vehicle and the characteristics of surrounding agents, an interactive prediction-decision-making planning model is constructed. This model can extract the two-way interaction characteristics between the vehicle and surrounding agents and predict the future trajectories of the surrounding agents. An adaptive trajectory generation model is constructed, combining the influences of safety, efficiency, and comfort. It can adaptively generate more human-like motion planning trajectories based on different environmental information to ensure driving safety.

[0058] The method specifically comprises the following steps:

[0059] S1: Construct a polynomial parameterized trajectory generation model that satisfies the vehicle's kinematic constraints and generates a set of future driving actions based on environmental information and the vehicle's current position.

[0060] Constructing a polynomial parameterized trajectory generation model specifically includes the following steps:

[0061] S11: Obtain local information of the vehicle: According to the current position of the vehicle and the high-precision map information provided, obtain the current motion state P of the vehicle c And the future drivable lane path point R, namely:

[0062] P c =[x c ,y c ,θ c ,v c ,a c ]

[0063] R=[L1,L2,...,L n ]

[0064]

[0065] Among them, x c ,y c ,θ c ,v c ,a c Respectively represent the horizontal and vertical coordinates, heading angle, speed and acceleration of the current position of the vehicle, L n and They represent the nth future drivable lane and the hth path point information of the lane, They represent the horizontal and vertical coordinates, navigation angle, curvature and speed limit of the h-th path point in the n-th lane respectively.

[0066] S12: Constructing a polynomial parameterized trajectory generation model: Based on the vehicle's current position information and future drivable path information provided in step S11, and in order to adapt to the complex structure of urban road scenes, a trajectory model of lateral displacement and longitudinal displacement versus time is constructed using a quintic polynomial and a quartic polynomial in the Frenet coordinate system, respectively:

[0067] Δs(t)=b0+b1t+b2t 2 +b3t 3 +b4t 4

[0068] Δd(t)=r0+r1t+r2t 2 +r3t 3 +r4t 4 +r5t 5

[0069] Among them, Δd(t) and Δs(t) represent the lateral and longitudinal displacements at time t, respectively. [b0, b1, ..., b4] and [r0, r1, ..., r4] are the parameters of the polynomial parameterized model, which can be obtained by the current position information P of the vehicle. c and the terminal position information P e Solve, P e The calculation can be performed by assuming different driving speeds. Finally, the obtained Δd(t) and Δs(t) are converted to the Cartesian coordinate system to obtain the final planned trajectory set

[0070] S2: Build an interactive prediction-decision planning model based on deep learning. Based on the two-way interaction between the autonomous vehicle and surrounding vehicles, extract the hidden features of the interaction and predict the trajectories of surrounding vehicles. The specific steps include:

[0071] S21: Interactive prediction-decision planning model input: The input of the interactive prediction-decision planning model mainly includes the historical trajectory characteristics of surrounding vehicles The local map information where the vehicle is currently located And the vehicle planning trajectory set obtained in step S12

[0072]

[0073] Among them, X i and They are respectively represented as the feature information and position features of the i-th intelligent agent, i.e., the vehicle, t h Represents the historical time domain, l i ,w i Respectively represent the horizontal and vertical coordinates, heading angle, horizontal and vertical speed, vehicle length and width at time t, Mn Represents the nth local lane information, They are respectively represented as the horizontal and vertical coordinates, navigation angle, lane type, speed limit and traffic light status from the first point to the jth path point.

[0074] S22: Input feature encoding: Encode the agent features obtained in step S21 using the LSTM network to obtain hidden features Map feature information is encoded using the MLP network to obtain hidden features The planned trajectory set is encoded using another MLP network to obtain hidden features

[0075] S23: Global feature encoding: In order to obtain the global feature information of the agent and the local map, the encoding results of the agent and map obtained in step S22 are encoded using the self-attention mechanism of the two-layer Transformer for global feature encoding:

[0076] G featrue =Attention(Q F ,K F ,V F ,mask)

[0077]

[0078] G feature ={G agent ,G map}

[0079] Among them, G featrue is a global feature, which can be decomposed into the agent global feature G agent and the global map feature G map , Q F ,K F ,V F ,mask are the query,key, value and corresponding mask of the attention mechanism, respectively.,Attention,and,concatenate,represent the attention mechanism and concatenation, respectively.

[0080] S24: Interactive feature information modeling: Combine the hidden features of the planning trajectory set obtained in step S22 The global feature G of the agent obtained in step S23 agent and the global map feature G map , use the cross attention mechanism to model the interaction of each predicted agent i:

[0081]

[0082] Among them, F i,am and Fi,ae They are respectively represented as the interaction features between the ith agent and the map and the planned trajectory, Q i,F , mask m , mask e They represent the query, key, value and corresponding mask required to calculate the attention respectively. Represents the global characteristics of the i-th agent.

[0083] S25: Prediction trajectory decoding: Combine the agent-map interaction feature F obtained in step S24 i,am and the interactive features F of the agent-planning trajectory i,ae , use an MLP to decode the future trajectory of the i-th agent:

[0084]

[0085] Among them, t f Indicates the future time domain.

[0086] S3: Build an adaptive trajectory generation model and a learnable trajectory selection module. Use the trajectory prediction results from step S2 to score the driving action set and combine the deviations to generate a network optimization result to ultimately generate the optimal trajectory. This specifically includes the following steps:

[0087] S31: Learnable cost weight: Combined with the hidden features of the vehicle's historical trajectory obtained in step S22 and the vehicle-map interactive hidden feature F obtained in step S24 e,am Splice and use an MLP network to decode the weight of the cost function

[0088]

[0089] Among them, τ s ,τ e ,τ c are expressed as the weights of safety, efficiency and comfort respectively.

[0090] S32: Trajectory scoring modeling: combining the weights obtained in S31 Build a trajectory scoring model:

[0091] U o =τ s U s +τ e U e +τ c U c

[0092]

[0093] Among them, U s ,U e ,U c denote the cost functions of safety, efficiency and comfort respectively, o is the index of the predicted agent in the safety cost function, S safety is the safety threshold, They represent the horizontal and vertical coordinates of the planned trajectory of the vehicle and the predicted trajectory of the agent o in the future, They represent the speed and lateral and longitudinal acceleration of the vehicle’s planned trajectory in the future, v limit Indicates the lane speed limit.

[0094] S33: Trajectory deviation optimization network: Combine the planned trajectory scores obtained in step S32 and select the trajectory with the highest score As the basic trajectory of the trajectory deviation optimization network, find the corresponding hidden features of the ego-vehicle planning trajectory Combined with the hidden features of the vehicle's historical trajectory obtained in step S22 and the vehicle-map interactive hidden feature F obtained in step S24 e,am , the coordinate deviation of the trajectory is decoded using the MLP network and added to the trajectory to generate the optimal trajectory.

[0095] S34: Loss function formulation: In order to make trajectory prediction more accurate and the driving behavior of the vehicle more human-like, a loss function is designed

[0096]

[0097] in, They are respectively represented as the prediction trajectory imitation loss, the planning trajectory imitation loss and the planning selection score loss, λ1, λ2, λ3 represent the weights corresponding to each loss, G i ,G e Represent the true trajectories of the predicted agent i and the ego vehicle, represents the planning trajectory obtained by the model, smoothhl1 and softmax are smooth L1 loss and activation functions respectively, y z is the planning trajectory score, a and Z are the number of predicted agents and the number of planned trajectories respectively.

[0098] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not limiting. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention can be modified or replaced by equivalents without departing from the purpose and scope of the technical solutions, which should all be included in the scope of the claims of the present invention.

Claims

1. A collaborative motion planning method for autonomous driving integrating interactive trajectory prediction, characterized in that: The method specifically comprises the following steps: S1: Construct a polynomial parameterized trajectory generation model that satisfies the vehicle's kinematic constraints and generates a set of future driving actions based on environmental information and the vehicle's current position. S2: Build an interactive prediction-decision planning model based on deep learning. Based on the two-way interaction between the autonomous vehicle and surrounding vehicles, extract the hidden features of the interaction and predict the trajectories of surrounding vehicles. The specific steps include: S21: Interactive prediction-decision planning model input: The input of the interactive prediction-decision planning model includes the historical trajectory characteristics of surrounding vehicles The local map information where the vehicle is currently located And the vehicle planning trajectory set Among them, X i and They are respectively represented as the feature information and position features of the i-th intelligent agent, namely the vehicle, t h Represents the historical time domain, l i ,w i Respectively represent the horizontal coordinate, vertical coordinate, heading angle, lateral speed, longitudinal speed, vehicle length and width at time t, M n Represents the nth local lane information, They are represented as the horizontal coordinate, vertical coordinate, navigation angle, lane type, speed limit and traffic light status from the first waypoint to the jth waypoint respectively; S22: Input feature encoding: Encode the agent features obtained in step S21 using the LSTM network to obtain hidden features Map feature information is encoded using the MLP network to obtain hidden features The planned trajectory set is encoded using another MLP network to obtain hidden features S23: Global feature encoding: In order to obtain the global feature information of the agent and the local map, the encoding results of the agent and map obtained in step S22 are encoded with global features using the self-attention mechanism of the two-layer Transformer; G featrue =Attention(Q F ,K F ,V F ,mask) G feature ={G agent ,G map } Among them, G featrue is a global feature, which can be decomposed into the agent global feature G agent and the global map feature G map , Q F ,K F ,V F ,mask are the query,key, value and corresponding mask of the attention mechanism, respectively. Attention and concatenate represent the attention mechanism and concatenation, respectively; S24: Interactive feature information modeling: Combine the hidden features of the planning trajectory set obtained in step S22 The global feature G of the agent obtained in step S23 agent and the global map feature G map , use the cross attention mechanism to model the interaction of each predicted agent i: Among them, F i,am and F i,ae They are respectively represented as the interaction features between the ith agent and the map and the planned trajectory, Q i,F , mask m , mask e Represent the query, key, value and corresponding mask required to calculate attention, respectively. Represents the global characteristics of the i-th agent; S25: Prediction trajectory decoding: Combine the agent-map interaction feature F obtained in step S24 i,am and the interactive features F of the agent-planning trajectory i,ae , using an MLP network to decode the future trajectory of the i-th agent Among them, t f Indicates the future time domain; S3: Build an adaptive trajectory generation model and a learnable trajectory selection module. Use the trajectory prediction results of step S2 to score the driving action set, and combine the deviation to generate network optimization results to finally generate the optimal trajectory.

2. The autonomous driving collaborative motion planning method according to claim 1, characterized in that: In step S1, a polynomial parameterized trajectory generation model is constructed, which specifically includes the following steps: S11: Obtain local information of the vehicle: According to the current position of the vehicle and the high-precision map information provided, obtain the current motion state P of the vehicle c And the future drivable lane path point R, namely: P c =[x c ,y c ,i c ,v c ,a c ] R=[L1,L2,...,L n ] Among them, x c ,y c ,θ c ,v c ,a c They represent the horizontal coordinate, vertical coordinate, heading angle, speed and acceleration of the current position of the vehicle, respectively. n and They represent the nth future drivable lane and the hth path point information of the lane, The horizontal coordinate, vertical coordinate, navigation angle, curvature and speed limit of the h-th path point in the n-th lane are represented respectively; S12: Constructing a polynomial parameterized trajectory generation model: Based on the current vehicle position information and the future drivable path information provided in step S11, a trajectory model of lateral displacement and longitudinal displacement versus time is constructed using a quintic polynomial and a quartic polynomial in the Frenet coordinate system: Δs(t)=b0+b1t+b2t 2 +b3t 3 +b4t 4 Δd(t)=r0+r1t+r2t 2 +r3t 3 +r4t 4 +r5t 5 Among them, Δd(t) and Δs(t) represent the lateral and longitudinal displacements at time t, respectively. [b0, b1, ..., b4] and [r0, r1, ..., r4] are the parameters of the polynomial parameterized model. The current position information P of the vehicle is used to calculate the c and the terminal position information P e Solve, P e The calculation is performed by assuming different driving speeds; finally, the obtained Δd(t) and Δs(t) are converted to the Cartesian coordinate system to obtain the final planned trajectory set 3. The autonomous driving collaborative motion planning method according to claim 2, characterized in that: In step S3, an adaptive trajectory generation model is constructed, which specifically includes the following steps: S31: Learnable cost weight: Combined with the hidden features of the vehicle's historical trajectory obtained in step S22 and the vehicle-map interactive hidden feature F obtained in step S24 e,am Splice and use an MLP network to decode the weight of the cost function Among them, τ s ,τ e ,τ c are expressed as the weights of safety, efficiency and comfort respectively; S32: Trajectory scoring modeling: Combine the weights obtained in step S31 Construct trajectory scoring model; U o =t s U s +t e U e +t c U c Among them, U o represents the trajectory score, U s ,U e ,U c denote the cost functions of safety, efficiency and comfort respectively, o is the index of the predicted agent in the safety cost function, S safety is the safety threshold, They represent the horizontal and vertical coordinates of the planned trajectory of the vehicle and the predicted trajectory of the agent o in the future, They represent the speed, lateral and longitudinal acceleration of the vehicle’s planned trajectory in the future, v limit Indicates lane speed limit; S33: Trajectory deviation optimization network: Combine the planned trajectory scores obtained in step S32 and select the trajectory with the highest score As the basic trajectory of the trajectory deviation optimization network, find the corresponding hidden features of the ego-vehicle planning trajectory Combined with the hidden features of the vehicle's historical trajectory obtained in step S22 and the vehicle-map interactive hidden feature F obtained in step S24 e,am , use the MLP network to decode the coordinate deviation of the trajectory and add it to the trajectory to generate the optimal trajectory; S34: Loss Function Formulation: Designing Loss Functions in, They represent the prediction trajectory imitation loss, planning trajectory imitation loss and planning selection score loss respectively, λ1, λ2, λ3 represent the weights corresponding to each loss, G i ,G e Represent the true trajectories of the predicted agent i and the ego vehicle, represents the planning trajectory obtained by the model, smoothhl1 and softmax are smooth L1 loss and activation functions respectively, y z is the planning trajectory score, a and Z are the number of predicted agents and the number of planned trajectories respectively.

Images