Methods, devices, equipment, storage media, and products for dynamically allocating driving rights coefficients.

Abstract

This application relates to a method, apparatus, computer device, storage medium, and computer program product for dynamically allocating driving rights coefficients. By establishing a vehicle path tracking dynamics model of the target vehicle and obtaining a first evaluation result corresponding to the driver's physiological state and a second evaluation result corresponding to trajectory tracking performance, and provided that the first evaluation result does not exceed a first threshold and the second evaluation result does not exceed a second threshold, a first driving rights coefficient for the driver is calculated based on the first and second evaluation indicators. This coefficient is then used to control the steering torque input by the driver. Furthermore, based on the first driving rights coefficient, a second driving rights coefficient for the autonomous driving system is determined to control the steering torque input by the autonomous driving system. A robust controller is obtained based on the vehicle path tracking dynamics model, enabling driving control through the autonomous driving system even when the driver's physiological state is abnormal, achieving dynamic allocation of driving rights coefficients, and ensuring the stability of the driving system.

Description

Technical Field

[0001] This application relates to the field of control algorithm technology, and in particular to a method, apparatus, computer device, storage medium and computer program product for dynamically allocating driving rights coefficients. Background Technology

[0002] With the development of autonomous driving technology, more and more commercial vehicles are beginning to adopt human-machine co-driving for driving control. Human-machine co-driving means that the task of driving the vehicle is jointly undertaken by the driver and the autonomous driving system.

[0003] In traditional methods, human-machine co-driving is usually based on a fixed driving rights allocation measurement, and the sum of the driving rights coefficients of the driver and the autonomous driving system is 1, that is, the autonomous driving system compensates for the insufficient input of the driver.

[0004] However, in some extreme cases, due to the saturation of the controller output of the autonomous driving system, the output of the human-machine co-driving system does not match the control input required to ensure vehicle driving safety, thereby reducing the reliability of the autonomous driving system. Furthermore, when there is a significant change in driving authority, it will affect the stability of the system. Summary of the Invention

[0005] Therefore, it is necessary to provide a method, device, computer equipment, computer-readable storage medium, and computer program product for dynamically allocating driving rights coefficients that can ensure the stability of the driving system, in order to address the above-mentioned technical problems.

[0006] Firstly, this application provides a method for dynamically allocating driving rights coefficients, applied to autonomous vehicles, the autonomous vehicles including an autonomous driving system; the method includes:

[0007] Establish a dynamic model for whole-vehicle path tracking of the target vehicle;

[0008] The first evaluation result corresponding to the driver's physiological state and the second evaluation result corresponding to the trajectory tracking performance are obtained respectively.

[0009] If the first evaluation result does not exceed the first threshold and the second evaluation result does not exceed the second threshold, the driver's first driving right coefficient is calculated based on the first evaluation index and the second evaluation index; the first driving right coefficient is used to control the steering torque input by the driver through the steering wheel;

[0010] Based on the first driving authority coefficient, the second driving authority coefficient of the autonomous driving system is determined; the second driving authority coefficient is used to control the steering torque input to the controller of the autonomous driving system; the sum of the first driving authority coefficient and the second driving authority coefficient is one.

[0011] A robust controller is obtained based on the vehicle path tracking dynamics model, and the driving rights coefficient is dynamically allocated through the robust controller.

[0012] In one embodiment, the step of establishing a vehicle path tracking dynamics model for the target vehicle includes:

[0013] Obtain the pre-established trajectory tracking model and lateral dynamics model of the target vehicle; the trajectory tracking model is used to represent the relative positional relationship between the target vehicle and the target trajectory during the trajectory tracking process;

[0014] Based on the lateral dynamics model, the driving parameters of the target vehicle are obtained;

[0015] Based on the relative positional relationship, the error parameters of the target vehicle are obtained; the error parameters include lateral position error and heading angle error.

[0016] The lateral deviation dynamic equation is established based on the lateral position error, and the angular deviation dynamic equation is established based on the heading angle error.

[0017] Based on driving parameters, lateral deviation dynamic equations, angular deviation dynamic equations, and preset variable parameters, a vehicle path tracking dynamic model for the target vehicle is established.

[0018] In one embodiment, the step of obtaining a first evaluation result corresponding to the driver's physiological state includes:

[0019] The driver's driving behavior images are continuously acquired according to the target frame number;

[0020] The system performs image recognition processing on images of driving behavior and determines the corresponding driving behavior of the driver based on the image recognition results. Driving behavior includes normal driving behavior, smoking, making a phone call, yawning, and driving with eyes closed.

[0021] For each type of driving behavior, a corresponding safety factor is set;

[0022] Based on all driving behaviors and corresponding safety factors, the first evaluation result corresponding to the driver's physiological state is calculated.

[0023] In one embodiment, the step of obtaining a second evaluation result corresponding to the trajectory tracking performance includes:

[0024] Based on the lateral position error, the second evaluation result corresponding to the trajectory tracking performance is calculated.

[0025] In one embodiment, the step of calculating the driver's first driving right coefficient based on a first evaluation index and a second evaluation index includes:

[0026] The first and second evaluation indicators were normalized respectively.

[0027] Based on the preset state levels, the first membership function corresponding to the normalized first evaluation index and the second membership function corresponding to the normalized second evaluation index are determined; the membership function is used to represent the membership degree of the corresponding evaluation index to each preset state level.

[0028] The driver's first driving rights coefficient is determined based on the first membership function and the second membership function.

[0029] In one embodiment, the step of obtaining a robust controller based on the vehicle path tracking dynamics model and dynamically allocating driving power coefficients using the robust controller includes:

[0030] The tire stiffness of the target vehicle is obtained based on the vehicle path tracking dynamics model, and a linear stiffness model is constructed based on the tire stiffness.

[0031] When tire stiffness changes, tire driving parameters are obtained, and a nonlinear stiffness model is constructed based on the tire driving parameters.

[0032] A fuzzy model is constructed based on linear stiffness models and nonlinear stiffness models;

[0033] With the goal of minimizing lateral position error and heading angle error, a robust controller is obtained based on a fuzzy model, a first driving authority coefficient, and a second driving authority coefficient. The driving authority coefficient is then dynamically allocated through the robust controller.

[0034] Secondly, this application also provides a dynamic allocation device for driving rights coefficients, the device comprising:

[0035] The model building module is used to build a dynamic model of the target vehicle's path tracking.

[0036] The result acquisition module is used to acquire the first evaluation result corresponding to the driver's physiological state and the second evaluation result corresponding to the trajectory tracking performance, respectively.

[0037] The coefficient calculation module is used to calculate the driver's first driving right coefficient based on the first evaluation index and the second evaluation index when the first evaluation result does not exceed the first threshold and the second evaluation result does not exceed the second threshold; the first driving right coefficient is used to control the steering torque input by the driver through the steering wheel;

[0038] The coefficient determination module is used to determine the second driving authority coefficient of the autonomous driving system based on the first driving authority coefficient; the second driving authority coefficient is used to control the steering torque input by the controller of the autonomous driving system; the sum of the first driving authority coefficient and the second driving authority coefficient is one.

[0039] The coefficient allocation module is used to obtain a robust controller based on the vehicle path tracking dynamics model, and to dynamically allocate driving rights coefficients through the robust controller.

[0040] Thirdly, this application also provides a computer device including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the method steps of any one of the first aspects.

[0041] Fourthly, this application also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the method steps of any one of the first aspects.

[0042] Fifthly, this application also provides a computer program product, which includes a computer program that, when executed by a processor, implements the method steps of any one of the first aspects.

[0043] The aforementioned method, device, computer equipment, storage medium, and computer program product for dynamically allocating driving rights coefficients establish a vehicle path tracking dynamics model of the target vehicle and obtain a first evaluation result corresponding to the driver's physiological state and a second evaluation result corresponding to the trajectory tracking performance. When the first evaluation result does not exceed a first threshold and the second evaluation result does not exceed a second threshold, the driver's first driving rights coefficient is calculated based on the first and second evaluation indicators. This first driving rights coefficient controls the steering torque input by the driver through the steering wheel. Furthermore, based on the first driving rights coefficient, a second driving rights coefficient for the autonomous driving system is determined to control the steering torque input by the controller of the autonomous driving system. A robust controller is obtained based on the vehicle path tracking dynamics model, enabling driving control through the autonomous driving system even when the driver's physiological state is abnormal. This achieves dynamic allocation of driving rights between the driver and the autonomous driving system, ensuring the stability of the driving system. Attached Figure Description

[0044] Figure 1 This is an application environment diagram of the dynamic allocation method for driving rights coefficients in one embodiment;

[0045] Figure 2 This is a flowchart illustrating the dynamic allocation method for driving rights coefficients in one embodiment;

[0046] Figure 3 This is a schematic diagram of the trajectory tracking process in one embodiment;

[0047] Figure 4 This is a schematic diagram of the membership function of the first evaluation index in one embodiment;

[0048] Figure 5 This is a schematic diagram of the membership function of the second evaluation index in one embodiment;

[0049] Figure 6 This is a schematic diagram of the membership function of the first driving authority coefficient in one embodiment;

[0050] Figure 7 This is a flowchart illustrating the dynamic allocation method for driving rights coefficients in one embodiment;

[0051] Figure 8 This is a structural block diagram of a dynamic allocation device for driving rights coefficients in one embodiment;

[0052] Figure 9 This is an internal structural diagram of a computer device in one embodiment. Detailed Implementation

[0053] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0054] The dynamic allocation method for driving rights coefficients provided in this application embodiment can be applied to, for example... Figure 1 In the application environment shown, terminal 102 communicates with controller 104 via a network. Terminal 102 acquires a pre-established trajectory tracking model and lateral dynamics model of the target vehicle, and based on these models, establishes a whole-vehicle path tracking dynamics model for the target vehicle. It then acquires a first evaluation result corresponding to the driver's physiological state and a second evaluation result corresponding to the trajectory tracking performance. If the first evaluation result does not exceed a first threshold and the second evaluation result does not exceed a second threshold, it calculates the driver's first driving right coefficient based on the first and second evaluation indicators. Based on the first driving right coefficient, it determines the second driving right coefficient of the autonomous driving system. The first driving right coefficient controls the steering torque input by the driver through the steering wheel; the second driving right coefficient controls the steering torque input by controller 104 of the autonomous driving system. Terminal 102 can be an in-vehicle terminal. Controller 104 is the controller of the autonomous driving system.

[0055] In one embodiment, such as Figure 2 As shown, a method for dynamically allocating driving rights coefficients is provided, which can be applied to... Figure 1 Taking terminal 102 as an example, the explanation includes the following steps:

[0056] S202: Establish a dynamic model for the whole vehicle path tracking of the target vehicle.

[0057] The vehicle path tracking dynamics model represents the dynamic equations of the target vehicle during trajectory tracking. During this process, the target vehicle is simultaneously subjected to steering torque input by the driver via the steering wheel and steering torque input by the autonomous driving system's controller. When establishing the vehicle path tracking dynamics model, the terminal controls the steering torque input by the steering wheel and controller by setting corresponding driving weight coefficients for both the driver and controller, thus preventing output saturation of the autonomous driving system. Furthermore, considering modeling errors and external interference, each variable in the vehicle path tracking dynamics model needs to be assigned a corresponding value, which represents the model uncertainty corresponding to each variable.

[0058] S204: Obtain the first evaluation result corresponding to the driver's physiological state and the second evaluation result corresponding to the trajectory tracking performance, respectively.

[0059] Since a driver's physiological state significantly impacts driving safety, the terminal needs to assess the driver's physiological state when determining the driver's driving rights coefficient. This assessment determines whether the driver is in a normal driving state, allowing the autonomous driving system to control the target vehicle when the driver is not driving normally. Specifically, the terminal collects driving images of the driver over a period of time, identifies and analyzes the driver's driving behavior, including normal driving, smoking, making phone calls, yawning, and closing eyes. In practical applications, more driving behaviors can be selected based on specific needs. Here, for ease of determining the first evaluation result, only the aforementioned driving behaviors are evaluated. For each driving behavior, the terminal determines the number of frames in the corresponding driving image and sets a corresponding safety coefficient. The first evaluation result corresponding to the driver's physiological state is calculated using the number of driving image frames and the corresponding driving coefficient. Regarding trajectory tracking performance, the terminal typically determines the second evaluation result based on the lateral position error between the current position coordinates and the corresponding position coordinates in the reference trajectory.

[0060] S206: If the first evaluation result does not exceed the first threshold and the second evaluation result does not exceed the second threshold, the driver's first driving right coefficient is calculated based on the first evaluation index and the second evaluation index; the first driving right coefficient is used to control the steering torque input by the driver through the steering wheel.

[0061] Before determining the driving rights coefficient, the terminal first makes a preliminary judgment on the first evaluation index and the second evaluation index. If the first evaluation result does not exceed the first threshold and the second evaluation result does not exceed the second threshold, it means that the driver's physiological state is approaching normal and will not affect the driving control of the target vehicle. Therefore, the terminal calculates the driver's first driving rights coefficient based on the first evaluation index and the second evaluation index to control the steering torque input by the driver through the steering wheel.

[0062] Because balancing the impact of multiple evaluation indicators on driving rights allocation is difficult, it's challenging to quantitatively determine the relationship between each indicator and the allocation coefficient. Therefore, a fuzzy logic algorithm is used to determine the driving rights coefficient. Specifically, the terminal normalizes the first and second evaluation indicators to the range [0,1]. For the first evaluation indicator, the terminal sets multiple driver physiological state levels; higher levels indicate a worse driving condition and a poorer autonomous driving control effect on the target vehicle, requiring more assistance from the autonomous driving system. For the second evaluation indicator, the terminal sets multiple trajectory tracking state levels; higher levels indicate a larger trajectory tracking error, also requiring assistance from the autonomous driving system.

[0063] In this evaluation, the physiological state level serves as the fuzzy set for the first evaluation index. The terminal uses a fuzzy algorithm to obtain the physiological state membership function of each fuzzy factor in the fuzzy set to which the first evaluation index belongs. Similarly, the trajectory tracking state level serves as the fuzzy set for the second evaluation index, and the terminal obtains the corresponding trajectory tracking state membership function. The physiological state membership function and the trajectory tracking state membership function are used together as inputs to the final fuzzy algorithm. The output of the fuzzy algorithm is the membership function of the first driving right coefficient, which the terminal then uses to determine the first driving right coefficient.

[0064] S208: Determine the second driving authority coefficient of the automated driving system based on the first driving authority coefficient; the second driving authority coefficient is used to control the steering torque input by the controller of the automated driving system; the sum of the first driving authority coefficient and the second driving authority coefficient is one.

[0065] In this case, since the sum of the first driving right coefficient and the second driving right coefficient is one when the first evaluation result does not exceed the first threshold and the second evaluation result does not exceed the second threshold, the terminal can directly determine the second driving right coefficient of the autonomous driving system after determining the first driving right coefficient, so as to control the steering torque input by the controller of the autonomous driving system.

[0066] S210: Obtain a robust controller based on the vehicle path tracking dynamics model, and dynamically allocate driving rights coefficients through the robust controller.

[0067] Robust controllers, in this context, refer to designs that enable a controlled object with uncertainties to achieve the desired control effect without requiring online parameter adjustments. In other words, the parameters of a robust controller are designed offline and remain unchanged during online operation. It ensures good control performance even when the controlled object's parameters fluctuate within a certain range. Specifically, when determining the robust controller, the terminal performs output feedback control simulation on the vehicle path tracking dynamics model. The optimal control quantity u is obtained through control, ensuring that the lateral position error and heading angle error of the target vehicle approach zero, thereby guaranteeing the target vehicle's trajectory tracking capability and resulting in optimal driving force coefficients for different driving scenarios.

[0068] In the aforementioned dynamic allocation method of driving rights coefficient, a vehicle path tracking dynamic model of the target vehicle is established, and a first evaluation result corresponding to the driver's physiological state and a second evaluation result corresponding to the trajectory tracking performance are obtained respectively. When the first evaluation result does not exceed a first threshold and the second evaluation result does not exceed a second threshold, the first driving rights coefficient of the driver is calculated according to the first evaluation index and the second evaluation index. The first driving rights coefficient is used to control the steering torque input by the driver through the steering wheel. Based on the first driving rights coefficient, the second driving rights coefficient of the autonomous driving system is determined to control the steering torque input by the controller of the autonomous driving system. A robust controller is obtained based on the vehicle path tracking dynamic model, which can realize driving control through the autonomous driving system when the driver's physiological state is abnormal, realize the dynamic allocation of driving rights between the driver and the autonomous driving system, and ensure the stability of the driving system.

[0069] In one embodiment, establishing a vehicle path tracking dynamics model for the target vehicle includes: acquiring a pre-established trajectory tracking model and a lateral dynamics model for the target vehicle; the trajectory tracking model is used to represent the relative positional relationship between the target vehicle and the target trajectory during trajectory tracking; obtaining the driving parameters of the target vehicle based on the lateral dynamics model; obtaining the error parameters of the target vehicle based on the relative positional relationship; the error parameters include lateral position error and heading angle error; establishing a lateral deviation dynamics equation based on the lateral position error, and establishing an angle deviation dynamics equation based on the heading angle error; and establishing a vehicle path tracking dynamics model for the target vehicle based on the driving parameters, the lateral deviation dynamics equation, the angle deviation dynamics equation, and the preset variable parameters.

[0070] The trajectory tracking model is derived from the relative positional relationship between the actual trajectory of the target vehicle and the reference trajectory during the trajectory tracking process. During the trajectory tracking process, the target vehicle moves along the pre-planned reference trajectory under the control of the autonomous driving system controller. However, in actual applications, due to the output saturation of the autonomous driving system controller, the driving trajectory of the target vehicle cannot completely match the reference trajectory, resulting in a deviation in the relative positional relationship between the two trajectories.

[0071] The lateral dynamics model is the dynamics model of the target vehicle in the lateral direction (perpendicular to the vehicle's direction of travel). During trajectory tracking, the target vehicle and the reference trajectory will generate lateral position error and heading angle error. The lateral position error is the lateral distance between the target vehicle's current position coordinates and the corresponding position coordinates in the reference trajectory, and the heading angle error is the angle difference between the target vehicle's current direction of travel and the ideal direction of travel at the corresponding position in the reference trajectory.

[0072] Among them, such as Figure 3 As shown, Figure 3 This represents the relative positional relationship between the target vehicle and the reference trajectory during trajectory tracking, where the curve represents the reference trajectory, and L = v x t p Let be the forward sight distance, where t p For the forward look time, v x For the lateral velocity, the terminal establishes a local coordinate system xoy and an inertial coordinate system XOY. During trajectory tracking, the target vehicle will generate a lateral position error e. y and heading angle error Among them, heading angle error The actual heading angle of the target vehicle With ideal heading angle The angle difference between them:

[0073] therefore, in, for The first derivative, for Given the first derivative of ρ, the curvature of the forward view point, and γ, the yaw rate, the lateral deviation dynamic equation of the target vehicle is: The dynamic equation for the angular deviation of the target vehicle is: in, For e y The first derivative, for The first derivative, v x Let v be the lateral velocity. y Let L be the longitudinal velocity, ρ be the curvature of the foresight point, and L = v. x tp Forward sight distance, t p γ is the forward look time, and γ is the yaw rate.

[0074] The target vehicle's lateral dynamics equations also include the target vehicle's longitudinal deviation dynamics equations and the target vehicle's yaw angle deviation dynamics equations.

[0075] The dynamic equation for longitudinal deviation is: The dynamic equation for yaw angle deviation is: in, For v y The first derivative, Let be the first derivative of γ, m be the mass, and v be the first derivative of γ. x For lateral velocity, I z Let C be the moment of inertia about the center of mass. f For the front wheel stiffness, C r For rear wheel stiffness, l f l is the distance from the center of mass to the front axle. r δ is the distance from the center of gravity to the rear axle, and δ is the steering torque.

[0076] Therefore, the constructed vehicle path tracking dynamics model can be expressed as: in:

[0077] u = [δ d ,δ c ] T w = [d1,d2,d3,d4,0,d5] T ,

[0078]

[0079] Where m is mass, e y This is the lateral position error. For the heading angle error, v x For lateral velocity, I z Let C be the moment of inertia about the center of mass. f For the front wheel stiffness, C r For rear wheel stiffness, l f l is the distance from the center of mass to the front axle. r v is the distance from the center of mass to the rear axle. x Let v be the lateral velocity. y ρ is the longitudinal velocity, γ is the curvature of the foresight point, γ is the yaw rate, and δ is the longitudinal velocity. f For an effective steering angle, δ d δ is the output steering angle for the driver. c Let δ be the output steering angle of the autonomous driving system controller, k1 be the driver's driving authority coefficient, and k2 be the controller's driving authority coefficient, where δf =k1δ d +k2δ c K s For the stiffness output by the autonomous driving system, I s C represents the moment of inertia output by the autonomous driving system. s The damping output of the autonomous driving system is given by d1, d2, d3, d4, and d5, which are the preset variable parameters corresponding to variable x, respectively. Where v x ρ is contained in d2.

[0080] In this embodiment, by acquiring the pre-established trajectory tracking model and lateral dynamics model of the target vehicle, the driving parameters of the target vehicle are obtained based on the lateral dynamics model, the error parameters of the target vehicle are obtained based on the relative position relationship, the lateral deviation dynamic equation is established based on the lateral position error, and the angular deviation dynamic equation is established based on the heading angle error. Based on the driving parameters, the lateral deviation dynamic equation, the angular deviation dynamic equation, and the preset variable parameters, the whole vehicle path tracking dynamic model of the target vehicle is established, which can realize the dynamic modeling of the path tracking of the target vehicle to simulate the actual operating state of the target vehicle.

[0081] In one embodiment, obtaining a first evaluation result corresponding to the driver's physiological state includes: continuously acquiring images of the driver's driving behavior according to a target number of frames; performing image recognition processing on the driving behavior images, and determining the driver's corresponding driving behavior based on the image recognition results; driving behavior includes normal driving behavior, smoking, making a phone call, yawning, and driving with eyes closed; setting a corresponding safety coefficient for each driving behavior; and calculating the first evaluation result corresponding to the driver's physiological state based on all driving behaviors and the corresponding safety coefficients.

[0082] In practical applications, the target frame count is typically 24 frames. The terminal continuously captures 24 frames of the driver's driving behavior video, obtaining a series of continuous driving behavior images. Image recognition processing is then performed on these images to determine whether the driver is engaging in normal driving, smoking, making a phone call, yawning, or driving with eyes closed, among other behaviors. The frame number corresponding to each driving behavior is recorded. A safety coefficient is set for each driving behavior, representing its impact on driving safety. This safety coefficient is then used to calculate the first evaluation result corresponding to the driver's physiological state. Specifically, the first evaluation result can be expressed as:

[0083]

[0084] Among them, J d As the first evaluation result, b max b is the maximum value of the safety factor.min f is the minimum safety factor. cl f is the frame rate corresponding to normal driving. sm f is the frame number corresponding to smoking. ya f is the number of frames corresponding to a phone call. ca f is the number of frames corresponding to a yawn. dr b is the frame number corresponding to driving with eyes closed. cl b represents the safety factor corresponding to normal driving. sm b represents the safety factor corresponding to smoking. ya b represents the security level corresponding to making a phone call. ca b is the safety factor corresponding to yawning. dr The safety factor corresponding to driving with eyes closed.

[0085] In this embodiment, by continuously acquiring images of the driver's driving behavior according to the target frame number, performing image recognition processing on the driving behavior images, and determining the corresponding driving behavior of the driver based on the image recognition results, a corresponding safety coefficient is set for each driving behavior, and a first evaluation result corresponding to the driver's physiological state is calculated based on all driving behaviors and the corresponding safety coefficients, which can accurately assess the driver's physiological state.

[0086] In one embodiment, the step of obtaining a second evaluation result corresponding to the trajectory tracking performance includes: calculating the second evaluation result corresponding to the trajectory tracking performance based on the lateral position error.

[0087] The lateral position error is determined based on the vehicle path tracking dynamics model, and the second evaluation result can be expressed as:

[0088]

[0089] Among them, J e As the second evaluation result, e y This is the lateral position error. This represents the heading angle error.

[0090] In this embodiment, by calculating the second evaluation result corresponding to the trajectory tracking performance based on the lateral position error, the trajectory tracking performance can be accurately evaluated.

[0091] In one embodiment, calculating the driver's first driving right coefficient based on a first evaluation index and a second evaluation index includes: normalizing the first evaluation index and the second evaluation index respectively; determining the first membership function corresponding to the normalized first evaluation index and the second membership function corresponding to the normalized second evaluation index based on preset state levels; the membership function is used to represent the membership degree of the corresponding evaluation index to each preset state level; and determining the driver's first driving right coefficient based on the first membership function and the second membership function.

[0092] Among them, the terminal respectively evaluates the first evaluation index J d Second evaluation index J e Normalization is performed to obtain the corresponding evaluation index J' normalized to the range [0,1]. d and J' e In terms of driver's physiological state, the terminal is set with corresponding status levels, including L1, L2, L3, L4, and L5. The higher the status level, the worse the driver's physiological state is, and the greater the impact on driving safety. In terms of trajectory tracking performance, the corresponding status levels include H1, H2, H3, H4, and H5. The higher the status level, the worse the trajectory tracking performance is, and the more likely the autonomous driving system is to assist driving.

[0093] Specifically, J' d Membership function such as Figure 4 As shown, Figure 4 The horizontal axis is J' d The value is shown on the vertical axis as J'. d Membership degree to the corresponding state level. J e The membership function of ' is as follows Figure 5 As shown, Figure 5 The horizontal axis is J e The value of ', with J on the vertical axis. e 'Membership degree to the corresponding state level.' d Membership function and J e The membership functions of ' are used as input parameters for the fuzzy algorithm, and the output of the fuzzy algorithm is the driver's initial first driving right coefficient k. 10 ,like Figure 6 As shown, Figure 6 The horizontal axis is k 10 The value of k is shown on the vertical axis. 10 Membership degree belonging to different driving states.

[0094] In some special cases, such as J e The value exceeds the physiological state threshold. And J d The value exceeds the trajectory tracking performance threshold. This indicates that the driver's physiological state is poor and the trajectory tracking performance of the current target vehicle is poor. In this case, the driver's primary driving authority coefficient needs to be set to 0.2, i.e., k1 = 0.2. Furthermore, the autonomous driving system is granted the highest level of driving authority, i.e., k2 = 1, to allow the autonomous driving system to take over driving control as much as possible. However, the driver retains some driving authority so that the driver can take over driving at any time. And in J... e The value did not exceed the physiological state threshold. And J d The value did not exceed the trajectory tracking performance threshold. In this case, it indicates that the driver's physiological state is good, and the automatic driving system participates in driving control as an auxiliary function. At this time, k1 = k 10 k2 = 1 - k1.

[0095] Additionally, in J e The value exceeds the physiological state threshold. And J d The value did not exceed the trajectory tracking performance threshold. In this case, it indicates that the driver's physiological state is poor, but the target vehicle's trajectory tracking performance is good. This means the driver's physiological state has little impact on driving safety, but the autonomous driving system still needs to take over driving control. Therefore, k1 = k 10 k2 = 1. In J e The value did not exceed the physiological state threshold. And J d The value exceeds the trajectory tracking performance threshold. In this situation, the driver's apparent physiological state is good, but the target vehicle's trajectory tracking performance is poor, requiring the intervention of the autonomous driving system for driving control. Therefore, k1 = k 10 k2 = 1.

[0096] In this embodiment, by normalizing the first evaluation index and the second evaluation index respectively, based on the preset state level, the first membership function corresponding to the normalized first evaluation index and the second membership function corresponding to the normalized second evaluation index are determined. Based on the first membership function and the second membership function, the driver's first driving right coefficient is determined. The first driving right coefficient can be accurately obtained, and when the driver's physiological state is abnormal, driving control is achieved through the automatic driving system to ensure the driving safety of the target vehicle.

[0097] In one embodiment, the step of obtaining a robust controller based on a vehicle path tracking dynamics model and dynamically allocating driving rights coefficients using the robust controller includes: obtaining the tire stiffness of the target vehicle based on the vehicle path tracking dynamics model and constructing a linear stiffness model based on the tire stiffness; obtaining tire driving parameters when the tire stiffness changes and constructing a nonlinear stiffness model based on the tire driving parameters; constructing a fuzzy model based on the linear stiffness model and the nonlinear stiffness model; and obtaining a robust controller based on the fuzzy model, a first driving rights coefficient, and a second driving rights coefficient, with the design objective of minimizing lateral position error and heading angle error, and dynamically allocating driving rights coefficients using the robust controller.

[0098] When the tire is in the linear region, its lateral stiffness is approximately constant. However, as road conditions and tire mass change, the tire lateral stiffness becomes a nonlinear function of tire normal force, road friction coefficient, tire slip ratio, and tire slip angle. Therefore, in the nonlinear region, the vehicle path tracking model is not a linear model. To control this nonlinear model, the terminal transforms the uncertain model containing tire stiffness and driving force coefficients into a fuzzy (TS) model, which is the sum of multiple linear models.

[0099] Specifically, the TS model can be represented as:

[0100]

[0101] in:

[0102]

[0103]

[0104] Where m is mass, e y This is the lateral position error. For the heading angle error, v x For lateral velocity, I z Let C be the moment of inertia about the center of mass. f For the front wheel stiffness, C r For rear wheel stiffness, l f l is the distance from the center of mass to the front axle. r v is the distance from the center of mass to the rear axle. x Let v be the lateral velocity. y ρ is the longitudinal velocity, γ is the curvature of the foresight point, γ is the yaw rate, and δ is the longitudinal velocity. f For an effective steering angle, δ d δ is the output steering angle for the driver. c Let δ be the output steering angle of the autonomous driving system controller, k1 be the driver's driving authority coefficient, and k2 be the controller's driving authority coefficient, where δ f=k1δ d +k2δ c K s For the stiffness output by the autonomous driving system, I s C represents the moment of inertia output by the autonomous driving system. s The damping output of the autonomous driving system. For the nonlinear function of front wheel stiffness, It is a nonlinear function of the rear wheel stiffness.

[0105] Assume that the change in velocity is divided into [v] xmin ,v xmax The driving rights allocation coefficients for the driver and controller vary within the range of [k]. 1min ,k 1max ]、[k 2min ,k 2max ], where v xmin For the minimum lateral velocity, v xmax k represents the maximum lateral velocity. 1min k is the minimum value of the first driving rights coefficient. 1max k represents the maximum value of the first driving rights coefficient. 2min k is the minimum value of the second driving rights coefficient. 2max This represents the maximum value of the second driving privilege coefficient.

[0106] We can obtain:

[0107]

[0108]

[0109] The weighting coefficients in the TS model are:

[0110] ζ1=h 11 h 21 h 31 h 41 ,ζ2=h 12 h 21 h 31 h 41 ,…,ζ 16 =h 12 h 22 h 32 h 42

[0111] Where ζ1,ζ2,…,ζ 16 h is the defined weighting coefficient. 11 ,h 12 ,h 31 ,h 41 ,h 12 ,h 22 ,h32 ,h 42 These are the control parameters for the defined controller.

[0112] In the robust controller, the control quantity of each subsystem is represented as: u i =-K i y = -K i Cx, where K is the gain matrix of each subsystem, then the total control quantity of the robust controller is expressed as:

[0113] i To minimize lateral position and heading angle errors, and considering parameter uncertainties and external disturbances, the gain matrix K is determined by finding the minimum value of the performance index function Ψ. i .

[0114] Wherein, Ψ can be represented as:

[0115] Where Q and R are weight matrices, usually selected based on practical experience, and used in solving for the gain matrix K. i When, the solution is obtained using linear matrix inequalities, such that K i The following inequalities must be satisfied:

[0116]

[0117] Among them, D i =(A 0i +ΔA i )P+B i N i Let P be a positive definite symmetric matrix, and solve for K. i =N i P is then used to determine the gain matrix of the robust controller, enabling dynamic allocation of driving rights coefficients through the robust controller.

[0118] In this embodiment, the tire stiffness of the target vehicle is obtained based on the vehicle path tracking dynamics model, and a linear stiffness model is constructed based on the tire stiffness. When the tire stiffness changes, tire driving parameters are obtained, and a nonlinear stiffness model is constructed based on the tire driving parameters. Based on the linear and nonlinear stiffness models, a fuzzy model is constructed with the goal of minimizing lateral position error and heading angle error. Based on the fuzzy model, the first driving power coefficient, and the second driving power coefficient, a robust controller is obtained. The driving power coefficient is dynamically allocated through the robust controller, which enables driving control through the autonomous driving system when the driver's physiological state is abnormal. This achieves dynamic allocation of driving power coefficients for the driver and the autonomous driving system, ensuring the stability of the driving system.

[0119] In one embodiment, such as Figure 7 As shown, a dynamic allocation method for driving rights coefficients is provided, applied to autonomous vehicles, which include an autonomous driving system; the method includes:

[0120] S702: Obtain the pre-established trajectory tracking model and lateral dynamics model of the target vehicle; the trajectory tracking model is used to represent the relative positional relationship between the target vehicle and the target trajectory during trajectory tracking; obtain the driving parameters of the target vehicle based on the lateral dynamics model; obtain the error parameters of the target vehicle based on the relative positional relationship; the error parameters include lateral position error and heading angle error; establish the lateral deviation dynamic equation based on the lateral position error, and establish the angular deviation dynamic equation based on the heading angle error; establish the whole vehicle path tracking dynamics model of the target vehicle based on the driving parameters, the lateral deviation dynamic equation, the angular deviation dynamic equation, and the preset variable parameters.

[0121] S704: Continuously acquire driver behavior images according to the target frame number; perform image recognition processing on the driving behavior images, and determine the corresponding driving behavior of the driver based on the image recognition results; driving behavior includes normal driving behavior, smoking, making a phone call, yawning, and driving with eyes closed; set a corresponding safety coefficient for each driving behavior; calculate the first evaluation result corresponding to the driver's physiological state based on all driving behaviors and the corresponding safety coefficients; calculate the second evaluation result corresponding to the trajectory tracking performance based on the lateral position error.

[0122] S706: If the first evaluation result does not exceed the first threshold and the second evaluation result does not exceed the second threshold, normalize the first evaluation index and the second evaluation index respectively; based on the preset state level, determine the first membership function corresponding to the normalized first evaluation index and the second membership function corresponding to the normalized second evaluation index; the membership function is used to represent the membership degree of the corresponding evaluation index to each preset state level; based on the first membership function and the second membership function, determine the driver's first driving right coefficient; the first driving right coefficient is used to control the steering torque input by the driver through the steering wheel.

[0123] S708: Determine the second driving authority coefficient of the automated driving system based on the first driving authority coefficient; the second driving authority coefficient is used to control the steering torque input by the controller of the automated driving system; the sum of the first driving authority coefficient and the second driving authority coefficient is one.

[0124] S710: Obtain the tire stiffness of the target vehicle based on the vehicle path tracking dynamics model, and construct a linear stiffness model based on the tire stiffness; when the tire stiffness changes, obtain the tire driving parameters, and construct a nonlinear stiffness model based on the tire driving parameters; construct a fuzzy model based on the linear stiffness model and the nonlinear stiffness model; with the minimum lateral position error and heading angle error as the design objective, obtain a robust controller based on the fuzzy model, the first driving force coefficient, and the second driving force coefficient, and dynamically allocate the driving force coefficient through the robust controller.

[0125] In this embodiment, a vehicle path tracking dynamics model of the target vehicle is established, and a first evaluation result corresponding to the driver's physiological state and a second evaluation result corresponding to the trajectory tracking performance are obtained respectively. If the first evaluation result does not exceed a first threshold and the second evaluation result does not exceed a second threshold, the driver's first driving right coefficient is calculated based on the first and second evaluation indicators. The first driving right coefficient is used to control the steering torque input by the driver through the steering wheel. Based on the first driving right coefficient, the second driving right coefficient of the autonomous driving system is determined to control the steering torque input by the controller of the autonomous driving system. A robust controller is obtained based on the vehicle path tracking dynamics model, which can realize driving control through the autonomous driving system when the driver's physiological state is abnormal, realize the dynamic allocation of the driving right coefficients of the driver and the autonomous driving system, and ensure the stability of the driving system.

[0126] It should be understood that although the steps in the flowcharts of the above embodiments are shown sequentially according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the flowcharts of the above embodiments may include multiple steps or multiple stages. These steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these steps or stages is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the steps or stages of other steps.

[0127] Based on the same inventive concept, this application also provides a dynamic allocation device for driving rights coefficients to implement the aforementioned dynamic allocation method. The solution provided by this device is similar to the implementation described in the above method; therefore, the specific limitations in one or more embodiments of the dynamic allocation device for driving rights coefficients provided below can be found in the limitations of the dynamic allocation method for driving rights coefficients described above, and will not be repeated here.

[0128] In one embodiment, such as Figure 8 As shown, a dynamic allocation device for driving rights coefficients is provided, comprising: a model building module 10, a result acquisition module 20, a coefficient calculation module 30, a coefficient determination module 40, and a coefficient allocation module 50, wherein:

[0129] Model building module 10 is used to build a whole vehicle path tracking dynamics model of the target vehicle based on the trajectory tracking model and the lateral dynamics model.

[0130] The result acquisition module 20 is used to acquire the first evaluation result corresponding to the driver's physiological state and the second evaluation result corresponding to the trajectory tracking performance, respectively.

[0131] The coefficient calculation module 30 is used to calculate the driver's first driving right coefficient based on the first evaluation index and the second evaluation index when the first evaluation result does not exceed the first threshold and the second evaluation result does not exceed the second threshold. The first driving right coefficient is used to control the steering torque input by the driver through the steering wheel.

[0132] The coefficient determination module 40 is used to determine the second driving authority coefficient of the autonomous driving system based on the first driving authority coefficient; the second driving authority coefficient is used to control the steering torque input by the controller of the autonomous driving system; the sum of the first driving authority coefficient and the second driving authority coefficient is one.

[0133] The coefficient allocation module 50 is used to obtain a robust controller based on the vehicle path tracking dynamics model, and to dynamically allocate the driving rights coefficients through the robust controller.

[0134] In one embodiment, the model building module 10 includes: a model acquisition unit, a parameter acquisition unit, an error acquisition unit, an equation building unit, and a model building unit, wherein:

[0135] The model acquisition unit is used to acquire the pre-established trajectory tracking model and lateral dynamics model of the target vehicle; the trajectory tracking model is used to represent the relative positional relationship between the target vehicle and the target trajectory during the trajectory tracking process;

[0136] The parameter acquisition unit is used to obtain the driving parameters of the target vehicle based on the lateral dynamics model.

[0137] The error acquisition unit is used to obtain the error parameters of the target vehicle based on the relative position relationship; the error parameters include lateral position error and heading angle error.

[0138] The equation-establishing unit is used to establish the lateral deviation dynamic equation based on the lateral position error and the angular deviation dynamic equation based on the heading angle error.

[0139] The model building unit is used to build a vehicle path tracking dynamic model of the target vehicle based on driving parameters, lateral deviation dynamic equations, angular deviation dynamic equations, and preset variable parameters.

[0140] In one embodiment, the result acquisition module 20 includes: an image acquisition unit, an image recognition unit, a coefficient setting unit, and a result acquisition unit, wherein:

[0141] The image acquisition unit is used to continuously acquire images of the driver's driving behavior according to the target number of frames.

[0142] The image recognition unit is used to perform image recognition processing on driving behavior images and determine the corresponding driving behavior of the driver based on the image recognition results; driving behavior includes normal driving behavior, smoking, making a phone call, yawning and driving with eyes closed.

[0143] The coefficient setting unit is used to set the corresponding safety coefficient for each driving behavior.

[0144] The result acquisition unit is used to calculate the first evaluation result corresponding to the driver's physiological state based on all driving behaviors and the corresponding safety coefficients.

[0145] In one embodiment, the result acquisition module 20 is further configured to calculate a second evaluation result corresponding to the trajectory tracking performance based on the lateral position error.

[0146] In one embodiment, the coefficient calculation module 30 includes: a normalization processing unit, a function determination unit, and a coefficient determination unit, wherein:

[0147] The normalization processing unit is used to normalize the first evaluation index and the second evaluation index respectively.

[0148] The function determination unit is used to determine the first membership function corresponding to the normalized first evaluation index and the second membership function corresponding to the normalized second evaluation index based on the preset state level; the membership function is used to represent the membership degree of the corresponding evaluation index to each preset state level.

[0149] The coefficient determination unit is used to determine the driver's first driving right coefficient based on the first membership function and the second membership function.

[0150] In one embodiment, the coefficient allocation module 50 includes: a stiffness acquisition unit, a model acquisition unit, a model construction unit, and a coefficient allocation unit, wherein:

[0151] The stiffness acquisition unit is used to obtain the tire stiffness of the target vehicle based on the vehicle path tracking dynamics model, and to construct a linear stiffness model based on the tire stiffness.

[0152] The model acquisition unit is used to acquire tire driving parameters when tire stiffness changes, and to construct a nonlinear stiffness model based on the tire driving parameters.

[0153] The model building unit is used to construct fuzzy models based on linear stiffness models and nonlinear stiffness models.

[0154] The coefficient allocation unit is designed to minimize lateral position error and heading angle error. Based on a fuzzy model, a first driving authority coefficient, and a second driving authority coefficient, a robust controller is obtained, and the driving authority coefficient is dynamically allocated through the robust controller.

[0155] Each module in the aforementioned dynamic allocation device for driving rights coefficients can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in or independent of the processor in a computer device, or stored in the memory of a computer device as software, so that the processor can call and execute the corresponding operations of each module.

[0156] In one embodiment, a computer device is provided, which may be a terminal, and its internal structure diagram may be as follows: Figure 9 As shown, the computer device includes a processor, memory, input / output interface, communication interface, display unit, and input device. The processor, memory, and input / output interface are connected via a system bus, and the communication interface, display unit, and input device are also connected to the system bus via the input / output interface. The processor provides computing and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage media. The input / output interface is used for exchanging information between the processor and external devices. The communication interface is used for wired or wireless communication with external terminals; wireless communication can be achieved through Wi-Fi, mobile cellular networks, NFC (Near Field Communication), or other technologies. When executed by the processor, the computer program implements a dynamic allocation method for driving rights coefficients. The display unit is used to form a visually visible image and can be a display screen, projection device, or virtual reality imaging device. The display screen can be an LCD screen or an e-ink screen. The input device of the computer device can be a touch layer covering the display screen, or buttons, trackballs, or touchpads set on the casing of the computer device, or external keyboards, touchpads, or mice, etc.

[0157] Those skilled in the art will understand that Figure 9The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0158] In one embodiment, a computer device is provided, including a memory and a processor. The memory stores a computer program, and the processor executes the computer program to perform the following steps: establishing a vehicle path tracking dynamics model of the target vehicle; obtaining a first evaluation result corresponding to the driver's physiological state and a second evaluation result corresponding to the trajectory tracking performance; calculating a first driving right coefficient of the driver based on the first evaluation index and the second evaluation index, provided that the first evaluation result does not exceed a first threshold and the second evaluation result does not exceed a second threshold; the first driving right coefficient is used to control the steering torque input by the driver through the steering wheel; determining a second driving right coefficient of the autonomous driving system based on the first driving right coefficient; the second driving right coefficient is used to control the steering torque input by the controller of the autonomous driving system; the sum of the first driving right coefficient and the second driving right coefficient is one; obtaining a robust controller based on the vehicle path tracking dynamics model, and dynamically allocating the driving right coefficients through the robust controller.

[0159] In one embodiment, the process of the processor executing the computer program involves establishing a vehicle path tracking dynamics model for the target vehicle, including: acquiring a pre-established trajectory tracking model and a lateral dynamics model of the target vehicle; the trajectory tracking model is used to represent the relative positional relationship between the target vehicle and the target trajectory during trajectory tracking; obtaining the driving parameters of the target vehicle based on the lateral dynamics model; obtaining the error parameters of the target vehicle based on the relative positional relationship; the error parameters include lateral position error and heading angle error; establishing a lateral deviation dynamics equation based on the lateral position error, and establishing an angle deviation dynamics equation based on the heading angle error; and establishing a vehicle path tracking dynamics model for the target vehicle based on the driving parameters, the lateral deviation dynamics equation, the angle deviation dynamics equation, and the preset variable parameters.

[0160] In one embodiment, the acquisition of a first evaluation result corresponding to the driver's physiological state when the processor executes the computer program includes: continuously acquiring images of the driver's driving behavior according to a target number of frames; performing image recognition processing on the driving behavior images and determining the corresponding driving behavior of the driver based on the image recognition results; driving behavior includes normal driving behavior, smoking, making a phone call, yawning, and driving with eyes closed; setting a corresponding safety coefficient for each driving behavior; and calculating the first evaluation result corresponding to the driver's physiological state based on all driving behaviors and the corresponding safety coefficients.

[0161] In one embodiment, the acquisition of a second evaluation result corresponding to trajectory tracking performance when the processor executes a computer program includes: calculating the second evaluation result corresponding to trajectory tracking performance based on the lateral position error.

[0162] In one embodiment, the calculation of the driver's first driving right coefficient based on a first evaluation index and a second evaluation index when the processor executes the computer program includes: normalizing the first evaluation index and the second evaluation index respectively; determining a first membership function corresponding to the normalized first evaluation index and a second membership function corresponding to the normalized second evaluation index based on preset state levels; the membership function is used to represent the membership degree of the corresponding evaluation index to each preset state level; and determining the driver's first driving right coefficient based on the first membership function and the second membership function.

[0163] In one embodiment, the processor executing a computer program involves obtaining a robust controller based on a vehicle path tracking dynamics model and dynamically allocating driving rights coefficients using the robust controller. This includes: obtaining the tire stiffness of the target vehicle based on the vehicle path tracking dynamics model and constructing a linear stiffness model based on the tire stiffness; obtaining tire driving parameters when the tire stiffness changes and constructing a nonlinear stiffness model based on the tire driving parameters; constructing a fuzzy model based on the linear stiffness model and the nonlinear stiffness model; and obtaining a robust controller based on the fuzzy model, a first driving rights coefficient, and a second driving rights coefficient, with the design objective of minimizing lateral position error and heading angle error, and dynamically allocating driving rights coefficients using the robust controller.

[0164] In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, it performs the following steps: establishing a vehicle path tracking dynamics model of the target vehicle; obtaining a first evaluation result corresponding to the driver's physiological state and a second evaluation result corresponding to the trajectory tracking performance; calculating a first driving right coefficient of the driver based on the first evaluation index and the second evaluation index, provided that the first evaluation result does not exceed a first threshold and the second evaluation result does not exceed a second threshold; the first driving right coefficient is used to control the steering torque input by the driver through the steering wheel; determining a second driving right coefficient of the autonomous driving system based on the first driving right coefficient; the second driving right coefficient is used to control the steering torque input by the controller of the autonomous driving system; the sum of the first driving right coefficient and the second driving right coefficient is one; obtaining a robust controller based on the vehicle path tracking dynamics model, and dynamically allocating the driving right coefficients through the robust controller.

[0165] In one embodiment, the computer program, when executed by a processor, involves establishing a vehicle path tracking dynamics model for the target vehicle, including: acquiring a pre-established trajectory tracking model and a lateral dynamics model of the target vehicle; the trajectory tracking model is used to represent the relative positional relationship between the target vehicle and the target trajectory during trajectory tracking; obtaining the driving parameters of the target vehicle based on the lateral dynamics model; obtaining the error parameters of the target vehicle based on the relative positional relationship; the error parameters include lateral position error and heading angle error; establishing a lateral deviation dynamics equation based on the lateral position error, and establishing an angle deviation dynamics equation based on the heading angle error; and establishing a vehicle path tracking dynamics model for the target vehicle based on the driving parameters, the lateral deviation dynamics equation, the angle deviation dynamics equation, and the preset variable parameters.

[0166] In one embodiment, the computer program, when executed by a processor, involves obtaining a first evaluation result corresponding to the driver's physiological state, including: continuously acquiring images of the driver's driving behavior according to a target number of frames; performing image recognition processing on the driving behavior images, and determining the driver's corresponding driving behavior based on the image recognition result; driving behavior includes normal driving behavior, smoking, making a phone call, yawning, and driving with eyes closed; setting a corresponding safety coefficient for each driving behavior; and calculating the first evaluation result corresponding to the driver's physiological state based on all driving behaviors and the corresponding safety coefficients.

[0167] In one embodiment, the acquisition of a second evaluation result corresponding to trajectory tracking performance when the computer program is executed by the processor includes: calculating the second evaluation result corresponding to trajectory tracking performance based on the lateral position error.

[0168] In one embodiment, when the computer program is executed by the processor, the calculation of the driver's first driving right coefficient based on a first evaluation index and a second evaluation index includes: normalizing the first evaluation index and the second evaluation index respectively; determining a first membership function corresponding to the normalized first evaluation index and a second membership function corresponding to the normalized second evaluation index based on preset state levels; the membership function is used to represent the membership degree of the corresponding evaluation index to each preset state level; and determining the driver's first driving right coefficient based on the first membership function and the second membership function.

[0169] In one embodiment, when the computer program is executed by the processor, it involves obtaining a robust controller based on the vehicle path tracking dynamics model and dynamically allocating driving rights coefficients through the robust controller. This includes: obtaining the tire stiffness of the target vehicle based on the vehicle path tracking dynamics model and constructing a linear stiffness model based on the tire stiffness; obtaining tire driving parameters when the tire stiffness changes and constructing a nonlinear stiffness model based on the tire driving parameters; constructing a fuzzy model based on the linear stiffness model and the nonlinear stiffness model; and obtaining a robust controller based on the fuzzy model, a first driving rights coefficient, and a second driving rights coefficient, with the design objective of minimizing lateral position error and heading angle error, and dynamically allocating driving rights coefficients through the robust controller.

[0170] In one embodiment, a computer program product is provided, including a computer program that, when executed by a processor, performs the following steps: establishing a vehicle path tracking dynamics model of the target vehicle; obtaining a first evaluation result corresponding to the driver's physiological state and a second evaluation result corresponding to the trajectory tracking performance; calculating a first driving right coefficient for the driver based on the first evaluation index and the second evaluation index, provided that the first evaluation result does not exceed a first threshold and the second evaluation result does not exceed a second threshold; the first driving right coefficient is used to control the steering torque input by the driver through the steering wheel; determining a second driving right coefficient for the autonomous driving system based on the first driving right coefficient; the second driving right coefficient is used to control the steering torque input by the controller of the autonomous driving system; the sum of the first driving right coefficient and the second driving right coefficient is one; obtaining a robust controller based on the vehicle path tracking dynamics model, and dynamically allocating the driving right coefficients through the robust controller.

[0171] In one embodiment, the computer program, when executed by a processor, involves establishing a vehicle path tracking dynamics model for the target vehicle, including: acquiring a pre-established trajectory tracking model and a lateral dynamics model of the target vehicle; the trajectory tracking model is used to represent the relative positional relationship between the target vehicle and the target trajectory during trajectory tracking; obtaining the driving parameters of the target vehicle based on the lateral dynamics model; obtaining the error parameters of the target vehicle based on the relative positional relationship; the error parameters include lateral position error and heading angle error; establishing a lateral deviation dynamics equation based on the lateral position error, and establishing an angle deviation dynamics equation based on the heading angle error; and establishing a vehicle path tracking dynamics model for the target vehicle based on the driving parameters, the lateral deviation dynamics equation, the angle deviation dynamics equation, and the preset variable parameters.

[0172] In one embodiment, the computer program, when executed by a processor, involves obtaining a first evaluation result corresponding to the driver's physiological state, including: continuously acquiring images of the driver's driving behavior according to a target number of frames; performing image recognition processing on the driving behavior images, and determining the driver's corresponding driving behavior based on the image recognition result; driving behavior includes normal driving behavior, smoking, making a phone call, yawning, and driving with eyes closed; setting a corresponding safety coefficient for each driving behavior; and calculating the first evaluation result corresponding to the driver's physiological state based on all driving behaviors and the corresponding safety coefficients.

[0173] In one embodiment, the acquisition of a second evaluation result corresponding to trajectory tracking performance when the computer program is executed by the processor includes: calculating the second evaluation result corresponding to trajectory tracking performance based on the lateral position error.

[0174] In one embodiment, when the computer program is executed by the processor, the calculation of the driver's first driving right coefficient based on a first evaluation index and a second evaluation index includes: normalizing the first evaluation index and the second evaluation index respectively; determining a first membership function corresponding to the normalized first evaluation index and a second membership function corresponding to the normalized second evaluation index based on preset state levels; the membership function is used to represent the membership degree of the corresponding evaluation index to each preset state level; and determining the driver's first driving right coefficient based on the first membership function and the second membership function.

[0175] In one embodiment, when the computer program is executed by the processor, it involves obtaining a robust controller based on the vehicle path tracking dynamics model and dynamically allocating driving rights coefficients through the robust controller. This includes: obtaining the tire stiffness of the target vehicle based on the vehicle path tracking dynamics model and constructing a linear stiffness model based on the tire stiffness; obtaining tire driving parameters when the tire stiffness changes and constructing a nonlinear stiffness model based on the tire driving parameters; constructing a fuzzy model based on the linear stiffness model and the nonlinear stiffness model; and obtaining a robust controller based on the fuzzy model, a first driving rights coefficient, and a second driving rights coefficient, with the design objective of minimizing lateral position error and heading angle error, and dynamically allocating driving rights coefficients through the robust controller.

[0176] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium, and when executed, it can include the processes of the embodiments of the above methods. Any references to memory, databases, or other media used in the embodiments provided in this application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM). The databases involved in the embodiments provided in this application may include at least one type of relational database and non-relational database. Non-relational databases may include, but are not limited to, blockchain-based distributed databases. The processors involved in the embodiments provided in this application may be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, quantum computing-based data processing logic devices, etc., and are not limited to these.

[0177] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0178] The embodiments described above are merely illustrative of several implementation methods of this application, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of this patent application. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this application should be determined by the appended claims.

Claims

1. A method for dynamically allocating driving rights coefficients, characterized in that, Applied to autonomous vehicles, the autonomous vehicles including an autonomous driving system; the method includes: Establish a dynamic model for whole-vehicle path tracking of the target vehicle; The first evaluation result corresponding to the driver's physiological state and the second evaluation result corresponding to the trajectory tracking performance are obtained respectively. If the first evaluation result does not exceed the first threshold and the second evaluation result does not exceed the second threshold, the first evaluation result and the second evaluation result are normalized respectively. Based on preset state levels, a first membership function corresponding to the first evaluation result after normalization and a second membership function corresponding to the second evaluation result after normalization are determined; the membership function is used to represent the membership degree of the corresponding evaluation result to each preset state level; Based on the first membership function and the second membership function, a first driving authority coefficient of the driver is determined; the first driving authority coefficient is used to control the steering torque input by the driver through the steering wheel; Based on the first driving authority coefficient, a second driving authority coefficient is determined for the autonomous driving system; the second driving authority coefficient is used to control the steering torque input by the controller of the autonomous driving system; the sum of the first driving authority coefficient and the second driving authority coefficient is one. A robust controller is obtained based on the vehicle path tracking dynamics model, and the driving authority coefficient is dynamically allocated through the robust controller.

2. The method according to claim 1, characterized in that, The establishment of the whole vehicle path tracking dynamics model for the target vehicle includes: Obtain a pre-established trajectory tracking model and lateral dynamics model of the target vehicle; the trajectory tracking model is used to represent the relative positional relationship between the target vehicle and the target trajectory during the trajectory tracking process; Based on the lateral dynamics model, the driving parameters of the target vehicle are obtained; Based on the relative positional relationship, the error parameters of the target vehicle are obtained; the error parameters include lateral position error and heading angle error. A lateral deviation dynamic equation is established based on the lateral position error, and an angular deviation dynamic equation is established based on the heading angle error. Based on the driving parameters, the lateral deviation dynamic equation, the angular deviation dynamic equation, and the preset variable parameters, a vehicle path tracking dynamic model for the target vehicle is established.

3. The method according to claim 1, characterized in that, The acquisition of the first evaluation result corresponding to the driver's physiological state includes: The driver's driving behavior images are continuously acquired according to the target frame number; The driving behavior images are subjected to image recognition processing, and the corresponding driving behavior of the driver is determined based on the image recognition results; the driving behavior includes normal driving behavior, smoking, making a phone call, yawning, and driving with eyes closed; For each type of driving behavior, a corresponding safety factor is set; Based on all driving behaviors and corresponding safety factors, the first evaluation result corresponding to the driver's physiological state is calculated.

4. The method according to claim 2, characterized in that, The second evaluation result corresponding to the trajectory tracking performance includes: Based on the lateral position error, a second evaluation result corresponding to the trajectory tracking performance is calculated.

5. The method according to claim 1, characterized in that, The step of obtaining a robust controller based on the vehicle path tracking dynamics model and dynamically allocating driving power coefficients through the robust controller includes: The tire stiffness of the target vehicle is obtained based on the vehicle path tracking dynamics model, and a linear stiffness model is constructed based on the tire stiffness. When the tire stiffness changes, tire driving parameters are obtained, and a nonlinear stiffness model is constructed based on the tire driving parameters. Based on the linear stiffness model and the nonlinear stiffness model, a fuzzy model is constructed; With the goal of minimizing lateral position error and heading angle error, a robust controller is obtained based on the fuzzy model, the first driving authority coefficient, and the second driving authority coefficient, and the driving authority coefficient is dynamically allocated through the robust controller.

6. A dynamic allocation device for driving rights coefficients, characterized in that, The device includes: The model building module is used to build a dynamic model of the target vehicle's path tracking. The result acquisition module is used to acquire the first evaluation result corresponding to the driver's physiological state and the second evaluation result corresponding to the trajectory tracking performance, respectively. The coefficient calculation module is used to normalize the first evaluation result and the second evaluation result respectively when the first evaluation result does not exceed the first threshold and the second evaluation result does not exceed the second threshold; based on preset state levels, it determines the first membership function corresponding to the normalized first evaluation result and the second membership function corresponding to the normalized second evaluation result; the membership function is used to represent the membership degree of the corresponding evaluation result to each preset state level; based on the first membership function and the second membership function, it determines the driver's first driving right coefficient; the first driving right coefficient is used to control the steering torque input by the driver through the steering wheel; The coefficient determination module is used to determine a second driving right coefficient for the autonomous driving system based on the first driving right coefficient; the second driving right coefficient is used to control the steering torque input by the controller of the autonomous driving system; the sum of the first driving right coefficient and the second driving right coefficient is one. The coefficient allocation module is used to obtain a robust controller based on the vehicle path tracking dynamics model, and to dynamically allocate driving rights coefficients through the robust controller.

7. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the method according to any one of claims 1 to 5.

8. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 5.

9. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 5.

Images