Personalized lane changing trajectory planning method and system based on risk prediction

Abstract

The invention discloses a personalized lane changing trajectory planning method and system based on risk prediction, and relates to the technical field of automatic driving, and the method comprises the steps: obtaining a natural driving data set, carrying out the preprocessing of the natural driving data set, carrying out the extraction of the preprocessed natural driving data set through a clustering algorithm, and obtaining the driving style data, the natural driving data set is acquired in an objective driving risk field environment based on a surrounding driving environment; obtaining driving style data based on a statistical analysis method to obtain risk prediction under each style, and determining differentiated safe drivable areas of different drivers based on the risk prediction under each style; and based on the differentiated safe drivable areas of different drivers, generating a vehicle driving track through a pre-established track planner considering dynamic constraints.

Description

Technical Field

[0001] The present invention relates to the field of autonomous driving technology, and in particular to a method and system for personalized lane change trajectory planning based on risk prediction. Background Art

[0002] Autonomous driving technology is an effective approach to addressing vehicle safety and traffic congestion. With the continuous development of autonomous driving technology, lane changing, as one of the key behaviors in the autonomous driving process, has attracted widespread attention. Currently, lane change planning is primarily based on the vehicle's environmental perception data and preset rules. However, these traditional methods mostly focus on path planning using the vehicle's sensors and environmental models, overlooking the impact of risk prediction on lane change decisions. Ignoring risk prediction can weaken the generalization ability of driving behavior models and improve ride comfort, thereby impacting the trust and user acceptance of intelligent driving systems. How to incorporate risk prediction into the decision-making process while considering objective driving environment information has become a major challenge in lane change trajectory planning. Summary of the Invention

[0003] In order to address the deficiencies mentioned in the above background technology, the purpose of the present invention is to provide a personalized lane change trajectory planning method and system based on risk prediction.

[0004] In a first aspect, the objectives of the present invention can be achieved by the following technical solution: a personalized lane change trajectory planning method based on risk prediction, the method comprising the following steps:

[0005] Acquiring a natural driving data set, preprocessing the natural driving data set, and extracting the preprocessed natural driving data set using a clustering algorithm to obtain driving style data, wherein the natural driving data set is collected based on an objective driving risk field environment of a surrounding driving environment;

[0006] Driving style data is acquired based on statistical analysis methods to obtain risk predictions for each style. Based on the risk predictions for each style, differentiated safe driving areas for different drivers are determined.

[0007] Based on the differentiated safe drivable areas of different drivers, the vehicle driving trajectory is generated through a pre-established trajectory planner that considers dynamic constraints.

[0008] In combination with the first aspect, in certain implementations of the first aspect, the method further includes: the objective driving risk field environment based on the surrounding driving environment is obtained through driving risk field theory.

[0009] In conjunction with the first aspect, in certain implementations of the first aspect, the method further includes: the process of acquiring the objective driving risk environment based on the surrounding driving environment includes:

[0010] The driving risk field expression of surrounding vehicles with different driving styles is as follows:

[0011]

[0012] in, represents the driving risk of the i-th surrounding vehicle, and denote the peak field intensity at the origin of the surrounding vehicle coordinate system and the vector at any point, φ x and φ y They are The angle between the X-axis and the Y-axis in the surrounding vehicle coordinate system, η is the lane line attenuation coefficient, when When the endpoint of is within the lane of the surrounding vehicle, η = 1; otherwise, η is a constant less than 1, and θ∈[0,1] represents the driving aggressiveness of the surrounding vehicle;

[0013] The expression for establishing the road driving risk field is as follows:

[0014]

[0015] Among them, y AV 、y f 、y r 、y c1 and y c2 Represent the lateral position of the autonomous vehicle, the left and right road boundary positions, and the left and right lane line positions; and σ j are the peak field intensity, safety distance parameter and scaling factor associated with each road feature j, respectively;

[0016] The final objective driving risk scene Its expression is as follows:

[0017]

[0018] in, and They represent the coordinates of the autonomous driving vehicle in the coordinate system of the i-th surrounding vehicle, thereby obtaining the objective driving risk field environment based on the surrounding driving environment.

[0019] In conjunction with the first aspect, in certain implementations of the first aspect, the method further includes: preprocessing the natural driving dataset, including:

[0020] Eliminate abnormal data and filter out lane change trajectories to obtain a processed natural driving dataset;

[0021] The processed natural driving data set is divided into three types of driving style data: aggressive, conservative and general using the Gaussian mixture clustering algorithm.

[0022] In conjunction with the first aspect, in certain implementations of the first aspect, the method further includes: the data of the three driving styles of aggressive, conservative, and general are expressed as follows:

[0023]

[0024] Where K represents the number of Gaussian components, π n represents the weight of the nth Gaussian component, Θ represents the parameter set of Gaussian mixture clustering, which can be expressed as Θ = {π m ,μ n ,Σ n}, p(χ m |Θ) represents the sample vector χ m The probability density function, N(χ m |μ n ,Σ n ) indicates that the mean vector is μ m And the covariance matrix is Σ n Gaussian distribution, d represents the dimension of the sample vector;

[0025] The data sets corresponding to the three types of driving styles, namely radical, conservative and general, are obtained as follows: a ,Ξ c and Ξ m , for Ξ a ,Ξ c ,Ξ m , and calculate the mean distance between the autonomous driving vehicle and the preceding vehicle at the end of the lane change using statistical analysis methods. And calculate the mean of the minimum distance between the autonomous driving vehicle and the following vehicle during the lane change process and the mean of its corresponding angle

[0026] according to as well as Combining formula (1), we can obtain the differentiated risk prediction boundaries of the driving environment for autonomous driving vehicles with different driving styles, and then obtain the differentiated safe driving areas for different drivers.

[0027] In combination with the first aspect, in certain implementations of the first aspect, the method further includes: the pre-established trajectory planner that takes into account dynamic constraints introduces a two-degree-of-freedom vehicle model and a linear tire model, and selects the state, input, and output variables of the longitudinal and lateral MPCs.

[0028] In conjunction with the first aspect, in certain implementations of the first aspect, the method further includes: the states, input and output variables of the longitudinal and transverse MPCs are as follows:

[0029]

[0030] Among them, ω is the yaw angle, δ is the front wheel turning angle, and then the time discrete model within the prediction range is constructed as follows:

[0031]

[0032] Among them, k x and k y are the time indices of longitudinal and lateral predictions, respectively, and t is the current time step. The objective function of the trajectory planner based on model predictive control is:

[0033]

[0034] Among them, N p is the prediction range, N c For the control horizon, the restricted model predictive control optimization problem is expressed as:

[0035]

[0036] in,

[0037] In conjunction with the first aspect, in certain implementations of the first aspect, the method further includes: planning the vehicle driving trajectory according to the following physical constraints:

[0038]

[0039] in, and are the minimum and maximum values of longitudinal acceleration respectively; and are the minimum and maximum values of the longitudinal acceleration change, respectively; δ min and δ max are the minimum and maximum values of the wheel angle respectively; Δδ min and Δδ max are the minimum and maximum values of the wheel angle changes respectively; α f,min and α f,max are the minimum and maximum values of the front wheel slip angle respectively; α r,min and α r,max Minimum and maximum values of the rear wheel slip angle.

[0040] In a second aspect, in order to achieve the above-mentioned objectives, the present invention discloses a personalized lane change trajectory planning system based on risk prediction, comprising:

[0041] a data processing module, configured to obtain a natural driving data set, preprocess the natural driving data set, and extract the preprocessed natural driving data set using a clustering algorithm to obtain driving style data, wherein the natural driving data set is collected based on an objective driving risk field environment of the surrounding driving environment;

[0042] A risk prediction module is used to obtain driving style data based on statistical analysis methods, obtain risk predictions for each style, and determine differentiated safe driving areas for different drivers based on the risk predictions for each style;

[0043] The trajectory planning module is used to generate the vehicle driving trajectory based on the differentiated safe driving areas of different drivers through a pre-established trajectory planner that considers dynamic constraints.

[0044] In another aspect of the present invention, in order to achieve the above-mentioned purpose, a terminal device is disclosed, including a memory, a processor, and a computer program stored in the memory and capable of running on the processor. The memory stores a computer program capable of running on the processor, and when the processor loads and executes the computer program, it adopts a personalized lane change trajectory planning method based on risk prediction as described above.

[0045] Beneficial effects of the present invention:

[0046] The present invention combines the driver's personalized needs with objective driving environment data, considers risk prediction during the trajectory planning process, and achieves more personalized lane change trajectory planning. Through cluster analysis methods, the driver's driving style is divided into three types: aggressive, conservative, and general. Differentiated safe drivable areas are set for drivers with different styles, significantly improving the accuracy of lane change trajectory planning. Through statistical analysis based on actual driving data and combined with objective risk field theory, the present invention can accurately extract risk predictions during the lane change process, which makes lane change trajectory planning more targeted and adaptable, avoiding the occurrence of unsafe driving. A trajectory planner that considers dynamic constraints is designed, and a model predictive control method is used for longitudinal and lateral trajectory planning. By introducing a two-degree-of-freedom vehicle model and a linear tire model, the computational complexity is reduced, and the accuracy and real-time performance of trajectory planning are effectively improved, ensuring the safety and comfort of the vehicle during the lane change process, thereby improving user acceptance of the intelligent driving system. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, those skilled in the art can derive other drawings based on these drawings without inventive effort.

[0048] Figure 1 It is a schematic flow chart of the method of the present invention;

[0049] Figure 2 This is a schematic diagram of the objective driving environment driving risk field of the present invention;

[0050] Figure 3 This is a schematic diagram of a safe drivable area based on differentiated risk predictions for different driving styles according to the present invention;

[0051] Figure 4 This is a schematic diagram of the personalized lane change trajectory planning effect based on subjective and objective consistent risk perception of the present invention;

[0052] Figure 5 It is a schematic diagram of the system structure of the present invention. DETAILED DESCRIPTION

[0053] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making any creative efforts shall fall within the scope of protection of the present invention.

[0054] Example 1:

[0055] like Figure 1 As shown, a personalized lane change trajectory planning method based on risk prediction includes the following steps:

[0056] S101: Obtain a natural driving dataset, preprocess the natural driving dataset, and extract the preprocessed natural driving dataset using a clustering algorithm to obtain driving style data, wherein the natural driving dataset is collected based on an objective driving risk field environment of a surrounding driving environment;

[0057] The objective driving risk field environment based on the surrounding driving environment is obtained through driving risk field theory.

[0058] The process of acquiring an objective driving risk field based on the surrounding driving environment involves: During lane changes, the autonomous vehicle's trajectory is significantly affected by road conditions and surrounding vehicles. Therefore, an objective driving risk field that accurately quantifies the driving environment is established. Specifically, the driving risk field for surrounding vehicles with different driving styles is expressed as follows:

[0059]

[0060] in, represents the driving risk of the i-th surrounding vehicle, and They represent the peak field intensity at the origin of the surrounding vehicle coordinate system and the vector at any point respectively. x and φ y They are The angle between the X-axis and the Y-axis in the surrounding vehicle coordinate system. η is the lane line attenuation coefficient. When the endpoint of lies within the lane of the surrounding vehicle, η = 1; otherwise, η is a constant much smaller than 1 because lane lines have a greater risk attenuation effect when the vehicle is traveling on a straight path. θ∈[0,1] represents the driving aggressiveness of the surrounding vehicle; values closer to 1 indicate a more aggressive driving style, while lower values indicate a more conservative driving style.

[0061] Furthermore, road boundaries and lane markings restrict the movement space of autonomous vehicles. When a vehicle approaches the edge of the road, the available space for obstacle avoidance is significantly reduced, increasing the risk of collision. However, approaching lane markings can lead to lane departure and vehicle-to-vehicle interference, threatening driving safety. Therefore, it is essential to establish a road driving risk field, which is expressed as follows:

[0062]

[0063] Among them, y AV 、y f 、y r 、y c1 and y c2 They represent the lateral position of the autonomous vehicle, the left and right road boundary positions, and the left and right lane line positions, respectively. and σ j are the peak field intensity, safety distance parameter, and scaling factor associated with each road feature j (whether it is a road boundary or a lane line).

[0064] The final objective driving risk scene Its expression is as follows:

[0065]

[0066] in, and Represent the coordinates of the autonomous driving vehicle in the coordinate system of the i-th surrounding vehicle. The objective driving risk field composed of the surrounding vehicles and the road is obtained as follows: Figure 2 shown.

[0067] S102: Acquire driving style data based on a statistical analysis method to obtain risk predictions for each driving style, and determine differentiated safe driving areas for different drivers based on the risk predictions for each driving style;

[0068] The preprocessing of the natural driving dataset includes:

[0069] Eliminate abnormal data and filter out lane change trajectories to obtain a processed natural driving dataset;

[0070] The processed natural driving data set is divided into three types of driving style data: aggressive, conservative and general using the Gaussian mixture clustering algorithm.

[0071]

[0072] Where K represents the number of Gaussian components, π n represents the weight of the nth Gaussian component, Θ represents the parameter set of Gaussian mixture clustering, which can be expressed as Θ = {π m ,μ n ,Σ n}, p(χ m |Θ) represents the sample vector χ m The probability density function, N(χ m |μ n ,Σ n ) indicates that the mean vector is μ m And the covariance matrix is Σ n Gaussian distribution, d represents the dimension of the sample vector.

[0073] Therefore, we can get the data sets corresponding to the three types of driving styles: aggressive, conservative and general, respectively. a ,Ξ c and Ξ m . Then, for Ξ a (or Ξ c , or m ) and calculate the mean distance between the autonomous vehicle and the preceding vehicle at the end of the lane change using statistical analysis methods. (or or ); and calculate the mean of the minimum distance between the autonomous vehicle and the following vehicle during the lane change process (or or ) and the mean of its corresponding angle (or or ).

[0074] according to (or or ), (or or )as well as (or or ), combined with formula (1), we can obtain the differentiated risk prediction boundaries of the driving environment for autonomous driving vehicles with different driving styles, and then obtain the differentiated safe driving areas for different drivers, such as Figure 3 shown.

[0075] S103: Based on the differentiated safe drivable areas of different drivers, a vehicle driving trajectory is generated through a pre-established trajectory planner that considers dynamic constraints.

[0076] To accommodate the varying longitudinal and lateral motion preferences of drivers with different driving styles, this paper designs separate model predictive controllers with different objectives for longitudinal and lateral trajectory planning. A two-degree-of-freedom vehicle model and a linear tire model are introduced to reduce the model's computational complexity. To accurately describe the vehicle's motion during lane changes, the following state, input, and output variables are selected for the longitudinal and lateral MPCs:

[0077]

[0078] Where ω is the yaw angle and δ is the front wheel turning angle. The time discrete model within the prediction range is then constructed as follows:

[0079]

[0080] Among them, k x and k y are the time indices of longitudinal and lateral predictions, respectively, and t is the current time step. The objective function of the trajectory planner based on model predictive control is:

[0081]

[0082] Among them, N p is the prediction range, N c is the control range. The restricted model predictive control optimization problem is expressed as:

[0083]

[0084] in, Trajectory planning can only be effectively implemented if the vehicle's physical constraints are met to ensure control stability and safety. The present invention establishes the following physical constraints to facilitate trajectory planning while maintaining the vehicle's dynamic response:

[0085]

[0086] in, and are the minimum and maximum values of longitudinal acceleration respectively; and are the minimum and maximum values of the longitudinal acceleration change, respectively; δ min and δ max are the minimum and maximum values of the wheel angle respectively; Δδ min and Δδ max are the minimum and maximum values of the wheel angle changes respectively; α f,min and α f,max are the minimum and maximum values of the front wheel slip angle respectively; α r,min and α r,max The minimum and maximum values of the side slip angles of the rear and front wheels. Finally, the results of personalized lane change trajectory planning based on subjective and objective consistent risk perception are as follows: Figure 4 As shown in the figure, personalized lane change trajectory planning can effectively achieve more refined trajectory adjustments based on differences in driving styles and risk perception. In particular, it can make optimized decisions based on risk prediction regarding obstacle avoidance and relative distance control to vehicles ahead and behind. Furthermore, the planning results demonstrate trajectory differences under different driving styles, demonstrating that this method can provide safe driving strategies that better meet individual needs based on the diversity of driving styles.

[0087] Example 2: The second aspect, as Figure 5 As shown, in order to achieve the above-mentioned purpose, the present invention discloses a personalized lane change trajectory planning system based on risk prediction, comprising:

[0088] A data processing module 11 is configured to obtain a natural driving data set, preprocess the natural driving data set, and extract the preprocessed natural driving data set using a clustering algorithm to obtain driving style data, wherein the natural driving data set is collected based on an objective driving risk field environment of the surrounding driving environment;

[0089] A risk prediction module 12 is configured to acquire driving style data based on a statistical analysis method, obtain risk predictions for each driving style, and determine differentiated safe driving areas for different drivers based on the risk predictions for each driving style;

[0090] The trajectory planning module 13 is used to generate a vehicle driving trajectory based on the differentiated safe driving areas of different drivers through a pre-established trajectory planner that considers dynamic constraints.

[0091] Based on the same inventive concept, the present invention also provides a computer device, which includes: one or more processors and a memory for storing one or more computer programs; the program includes program instructions, and the processor is used to execute the program instructions stored in the memory. The processor may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. It is the computing core and control core of the terminal, which is used to implement one or more instructions, specifically for loading and executing one or more instructions in a computer storage medium to implement the above method.

[0092] It should be further explained that, based on the same inventive concept, the present invention also provides a computer storage medium having a computer program stored thereon, which executes the above method when executed by a processor. The storage medium can be any combination of one or more computer-readable media. The computer-readable medium can be a computer-readable signal medium or a computer-readable storage medium. The computer-readable storage medium can be, for example, but not limited to, an electrical, magnetic, optical, electrical, magnetic, infrared, or semiconductor system, device or component, or any combination thereof. More specific examples (a non-exhaustive list) of computer-readable storage media include: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination thereof. In the present invention, a computer-readable storage medium can be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, device or component.

[0093] Throughout this specification, references to terms such as "one embodiment," "example," or "specific example" indicate that a specific feature, structure, material, or characteristic described in conjunction with that embodiment or example is included in at least one embodiment or example of the present disclosure. In this specification, schematic representations of these terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in any one or more embodiments or examples.

[0094] The above shows and describes the basic principles, main features and advantages of the present disclosure. Those skilled in the art should understand that the present disclosure is not limited to the above embodiments. The above embodiments and descriptions are merely illustrative of the principles of the present disclosure. Various changes and improvements may be made to the present disclosure without departing from the spirit and scope of the present disclosure, and such changes and improvements shall fall within the scope of the present disclosure.

Claims

1. A personalized lane change trajectory planning method based on risk prediction, characterized in that: The method comprises the following steps: Acquiring a natural driving data set, preprocessing the natural driving data set, and extracting the preprocessed natural driving data set using a clustering algorithm to obtain driving style data, wherein the natural driving data set is collected based on an objective driving risk field environment of a surrounding driving environment; Driving style data is acquired based on statistical analysis methods to obtain risk predictions for each style. Based on the risk predictions for each style, differentiated safe driving areas for different drivers are determined. Based on the differentiated safe drivable areas of different drivers, the vehicle driving trajectory is generated through a pre-established trajectory planner that considers dynamic constraints.

2. The personalized lane change trajectory planning method based on risk prediction according to claim 1, characterized in that: The objective driving risk field environment based on the surrounding driving environment is obtained through driving risk field theory.

3. The personalized lane change trajectory planning method based on risk prediction according to claim 2, characterized in that: The process of obtaining the objective driving risk environment based on the surrounding driving environment includes: The driving risk field expression of surrounding vehicles with different driving styles is as follows: in, represents the driving risk of the i-th surrounding vehicle, and denote the peak field intensity at the origin of the surrounding vehicle coordinate system and the vector at any point, φ x and φ y They are The angle between the X-axis and the Y-axis in the surrounding vehicle coordinate system, η is the lane line attenuation coefficient, when When the endpoint of is within the lane of the surrounding vehicle, η = 1; otherwise, η is a constant less than 1. Indicates the driving aggressiveness of surrounding vehicles; The expression for establishing the road driving risk field is as follows: Among them, y AV 、y f 、y r 、y c1 and y c2 Represent the lateral position of the autonomous vehicle, the left and right road boundary positions, and the left and right lane line positions; and σ j are the peak field intensity, safety distance parameter and scaling factor associated with each road feature j, respectively; The final objective driving risk scene Its expression is as follows: in, and They represent the coordinates of the autonomous driving vehicle in the coordinate system of the i-th surrounding vehicle, thereby obtaining the objective driving risk field environment based on the surrounding driving environment.

4. The personalized lane change trajectory planning method based on risk prediction according to claim 1, characterized in that: The preprocessing of the natural driving dataset includes: Eliminate abnormal data and filter out lane change trajectories to obtain a processed natural driving dataset; The processed natural driving data set is divided into three types of driving style data: aggressive, conservative and general using the Gaussian mixture clustering algorithm.

5. The personalized lane change trajectory planning method based on risk prediction according to claim 4 is characterized in that: The data expressions of the three driving styles of aggressive, conservative and general are as follows: Where K represents the number of Gaussian components, π n represents the weight of the nth Gaussian component, Θ represents the parameter set of Gaussian mixture clustering, which can be expressed as Θ = {π m ,μ n ,Σ n }, p(χ m |Θ) represents the sample vector χ m The probability density function, N(χ m |μ n ,Σ n ) indicates that the mean vector is μ m And the covariance matrix is Σ n Gaussian distribution, d represents the dimension of the sample vector; The data sets corresponding to the three types of driving styles, namely radical, conservative and general, are obtained as follows: a ,Ξ c and Ξ m , for Ξ a ,Ξ c ,Ξ m , and calculate the mean distance between the autonomous driving vehicle and the preceding vehicle at the end of the lane change using statistical analysis methods. And calculate the mean of the minimum distance between the autonomous driving vehicle and the following vehicle during the lane change process and the mean of its corresponding angle according to as well as Combining formula (1), we can obtain the differentiated risk prediction boundaries of the driving environment for autonomous driving vehicles with different driving styles, and then obtain the differentiated safe driving areas for different drivers.

6. The personalized lane change trajectory planning method based on risk prediction according to claim 1, characterized in that: The pre-established trajectory planner considering dynamic constraints introduces a two-degree-of-freedom vehicle model and a linear tire model, and selects the state, input and output variables of the longitudinal and lateral MPCs.

7. The personalized lane change trajectory planning method based on risk prediction according to claim 6, characterized in that: The state, input and output variables of the longitudinal and transverse MPCs are as follows: Among them, ω is the yaw angle, δ is the front wheel turning angle, and then the time discrete model within the prediction range is constructed as follows: Among them, k x and k y are the time indices of longitudinal and lateral predictions, respectively, and t is the current time step. The objective function of the trajectory planner based on model predictive control is: Among them, N p is the prediction range, ▽=yt+itp|t-yref,t+it p |t,N c For the control horizon, the restricted model predictive control optimization problem is expressed as: in, 8. The personalized lane change trajectory planning method based on risk prediction according to claim 1 is characterized in that: The vehicle trajectory planning must be performed according to the following physical constraints: in, and are the minimum and maximum values of longitudinal acceleration respectively; and are the minimum and maximum values of the longitudinal acceleration change, respectively; δ min and δ max are the minimum and maximum values of the wheel angle respectively; Δδ min and Δδ max are the minimum and maximum values of the wheel angle changes respectively; α f,min and α f,max are the minimum and maximum values of the front wheel slip angle respectively; α r,min and α r,max Minimum and maximum values of the rear wheel slip angle.

9. A personalized lane change trajectory planning system based on risk prediction, characterized in that: include: a data processing module, configured to obtain a natural driving data set, preprocess the natural driving data set, and extract the preprocessed natural driving data set using a clustering algorithm to obtain driving style data, wherein the natural driving data set is collected based on an objective driving risk field environment of the surrounding driving environment; A risk prediction module is used to obtain driving style data based on statistical analysis methods, obtain risk predictions for each style, and determine differentiated safe driving areas for different drivers based on the risk predictions for each style; The trajectory planning module is used to generate the vehicle driving trajectory based on the differentiated safe driving areas of different drivers through a pre-established trajectory planner that considers dynamic constraints.

10. A terminal device comprising a memory, a processor, and a computer program stored in the memory and capable of running on the processor, characterized in that: The memory stores a computer program that can be run on the processor. When the processor loads and executes the computer program, it adopts a personalized lane change trajectory planning method based on risk prediction according to any one of claims 1 to 8.

Images