Distributed driving electric vehicle path tracking and stability control method and system

Abstract

The invention discloses a distributed driving electric vehicle path tracking and stability control method and system, and belongs to the technical field of electric vehicle active safety control. The method comprises the following steps: constructing a state-space equation based on a driver path tracking model and a vehicle three-degree-of-freedom motion model; establishing a dynamic envelope region based on a phase plane method, inputting the current parameters of the vehicle into the dynamic stable envelope region, and outputting a weight; constructing a controller based on the state-space equation, and taking the weight output by the dynamic stable envelope region as the control weight of the controller; inputting the vehicle state variables and the vehicle parameters into a controller, and outputting a longitudinal driving torque, an additional yaw torque, an additional front wheel turning angle and a driver intention turning angle required by driving; the additional front wheel turning angle and the driver intention turning angle are superposed to serve as input of a steering motor; a torque distributor is designed to distribute a longitudinal driving torque and an additional yaw torque, and the distributed torques are used as given values of the four hub motors.

Description

Technical Field

[0001] This invention belongs to the field of active safety control technology for electric vehicles, specifically relating to a method and system for path tracking and stability control of distributed drive electric vehicles. Background Technology

[0002] With the continuous development of technology and the rapid growth of car ownership, environmental pollution and traffic safety issues are becoming increasingly prominent. Distributed drive electric vehicles (DDEVs), due to their structural advantages of independent in-wheel motor drive, can achieve precise and rapid drive control, and have become an important platform for improving vehicle driving performance and active safety control. Among them, active front steering (AFS) and direct yaw moment control (DYC), as two typical lateral control technologies, have been extensively studied in assisting drivers with path tracking and improving stability.

[0003] In existing technologies, Professor Liu Hui of Beijing University of Science and Technology proposed an AFS-based assisted steering system (Handling and Stability Integrated Control of AFS and DYC for Distributed Drive Electric Vehicles Based on Risk Assessment and Prediction), which improves vehicle lateral performance by introducing an additional steering angle within a human-machine shared framework. However, these methods mainly focus on the interaction between the driver and AFS, failing to consider the multi-actuator characteristics of the vehicle. Academician Guo Konghui's team at Jilin University, based on a DYC system, generates additional yaw moment through differential driving torque to improve vehicle stability under extreme conditions. Their work also designed a state observer to estimate the center-of-gravity sideslip angle, enhancing the practicality of DYC. Their further research integrates AFS and DYC, utilizing fuzzy control, robust control, or model predictive control methods to attempt to balance stability and handling under complex road conditions.

[0004] In addition, other studies have explored different control frameworks. For example, Professor Liang Zhongchao of Northeastern University uses a supervised system to trigger AFS or DYC control under specific operating conditions (Supervisor-Based Hierarchical Adaptive MPC for Yaw Stabilization of FWID-EVs Under Extreme Conditions); Associate Professor Chen Xiang of Nanjing University of Aeronautics and Astronautics (Robust LMI-Based H-Infinite Controller Integrating AFS and DYC of Autonomous Vehicles With Parametric Uncertainties) and Professor Yin Guodong of Southeast University (Enhancing Robust Driver Assistance Control in Distributed Drive Electric Vehicles through Integrated AFS and DYC Technology) use advanced methods such as gain scheduling, non-smooth control, and tubular model predictive control to improve system robustness. These studies provide multiple solutions for vehicle lateral control. However, existing solutions still have the following shortcomings:

[0005] Horizontal and vertical coupling problem: Most methods are based on a unified optimization model, with strong horizontal and vertical state coupling, making it difficult for the control strategy to simultaneously take into account both maneuverability and stability.

[0006] Single control objective: Traditional model predictive control (MPC) usually sets a uniform objective function, which means that AFS and DYC can only be optimized under the same index. The lack of interactive information and differentiated objectives limits the full play of their respective advantages.

[0007] Lack of adaptive coordination mechanism: Existing methods often assign fixed control roles to AFS and DYC, failing to dynamically adjust their cooperative relationship according to the vehicle's operating status, which can easily lead to resource waste or insufficient performance.

[0008] Therefore, existing technologies urgently need a new control method to achieve decoupling design of AFS and DYC, realize dual objective function optimization through interactive information, and dynamically adjust the coordination strategy based on the vehicle's operating status, thereby taking into account both vehicle handling and stability, and improving overall driving safety and comfort. Summary of the Invention

[0009] To address the shortcomings of existing technologies, the present invention aims to provide a method and system for path tracking and stability control of distributed drive electric vehicles, thereby solving the problems in the prior art.

[0010] The objective of this invention can be achieved through the following technical solutions:

[0011] This invention proposes a path tracking and stability control method for distributed drive electric vehicles, comprising the following steps:

[0012] A distributed drive electric vehicle path tracking and stability control method includes the following steps:

[0013] Based on the driver path tracking model and the vehicle three-degree-of-freedom motion model, a state-space equation is constructed.

[0014] A dynamic envelope region is established based on the phase plane method, and the current vehicle parameters are input into the dynamic stable envelope region, and the weights are output.

[0015] The controller is constructed based on the state-space equation, and the weights output from the dynamic stable envelope region are used as the control weights of the controller. Then, the vehicle state variables and vehicle parameters are input into the controller, and the outputs the longitudinal driving torque required for driving, the additional yaw torque, the additional front wheel steering angle, and the driver's intended steering angle. The additional front wheel steering angle is superimposed with the driver's intended steering angle as the input to the steering motor.

[0016] Design a torque distributor to distribute the longitudinal drive torque and additional yaw torque, and use the distributed torque as the setpoint for the four hub motors.

[0017] Furthermore, the driver path tracking model is as follows:

[0018]

[0019] Among them, L p e represents the driver's aiming distance. y This represents the lateral deviation between the vehicle's current position and the near-target point. v represents the heading error between the vehicle's current heading and the reference direction. y and v x These represent the vehicle's lateral velocity and longitudinal velocity, respectively; γ represents the vehicle's yaw rate; ρ r The curvature representing the reference path;

[0020] The three-degree-of-freedom motion model of the vehicle is as follows:

[0021]

[0022] F yf =-C f α f F yr =-C r α r

[0023] α f =β+l f r / v x -δα r =β-l r r / v x

[0024] Where m represents the total mass of the vehicle, v x β represents the vehicle's longitudinal velocity, r represents the vehicle's sideslip angle, and F represents the vehicle's yaw rate. xfl F xfr F xrl F xrr F represents the longitudinal force acting on the left front wheel, right front wheel, left and right wheels, and right rear wheel, respectively. yfl F yfr F yrl F yrr F represents the lateral force F acting on the left front wheel, right front wheel, left and right wheels, and right rear wheel, respectively. yf F yr C represents the lateral forces acting on the front and rear axles of the vehicle, respectively. f and C r These represent the lateral stiffness of the front and rear axles of the vehicle, respectively; I z The moment of inertia of a vehicle, l f l r These represent the distances between the front and rear axles and the vehicle's center of gravity, respectively; δ represents the front wheel steering angle, determined by the driver's intended steering angle δ. driver The additional front wheel steering angle δ calculated by the AFS system AFS The result is obtained by superposition; ΔM represents the additional yaw torque; α f and α r These represent the slip angles of the front and rear wheels, respectively.

[0025] Furthermore, the state-space equation is:

[0026]

[0027] Where, u = [F x ,δ AFS ,ΔM] T ,d=[δ driver ,ρ r ] T ,

[0028]

[0029]

[0030] Where u is the control input, F xLet d be the longitudinal force required for wheel drive, and A be the system disturbance quantity. t Let B be the state transition matrix of the system. t Let E be the input transition matrix of the system. t Let X be the system's disturbance transfer matrix; X is the vehicle state variable.

[0031] Furthermore, the envelope region is a quadrilateral with four vertices (r). C ,β C ),(r D ,β D ),(-r C ,-β C ),(-r D ,-β D The calculation formula for ) is as follows:

[0032]

[0033] Where, α sl,f Indicates the saturation value of the front wheel slip angle, α sl,r δ represents the saturation value of the rear wheel slip angle. max denoted as the maximum steering angle of the front wheel, μ as the road adhesion coefficient, and g as the acceleration due to gravity.

[0034] Furthermore, the controller includes: a drive agent, a front wheel steering agent, and a yaw torque agent; the drive agent constitutes a longitudinal control system to calculate the total longitudinal driving force required to track the longitudinal speed of the vehicle; the front wheel steering agent and the yaw torque agent constitute a lateral control system, with the control objectives being path tracking and lateral stability.

[0035] The objective cost functions for each agent are as follows:

[0036]

[0037]

[0038] In the formula, J1, J2, and J3 represent the cost functions of the driving agent, the front wheel steering agent, and the yaw torque agent, respectively.

[0039] The constraints include:

[0040] u min ≤u(k+j)≤u max

[0041] Δu min ≤Δu(k+j)≤Δu max

[0042] γ min (k+i)≤γ(k+i)≤γ max (k+i)

[0043] α rmin (k+i)≤α r (k+i)≤α rmax (k+i)

[0044] Where, N c Let u(k+j) and Δu(k+j) represent the input control value and input control increment at step k, respectively. max and u min These represent the maximum and minimum input values, Δu, respectively. max and Δu min Representing the maximum and minimum values ​​of the input increment, γ(k+i) and α, respectively. r (k+i) represent the yaw rate and the sideslip angle at the i-th step, respectively, γ max (k+i) and γ min (k+i) represent the maximum and minimum values ​​of the yaw rate, respectively, α rmax (k+i) and α rmin (k+i) represent the maximum and minimum values ​​of the centroid sideslip angle, respectively.

[0045] Furthermore, the cost function of the torque distributor is:

[0046]

[0047] F xi r w =T i ,i=fl,fr,rl,rr

[0048] Among them, J torque The cost function of the torque distributor. λ represents the average longitudinal attachment utilization rate; w1 and w2 are weighting factors; λ i To define the longitudinal attachment utilization rate, F xi F zi Let fl, fr, rl, rr represent the longitudinal force and vertical load of the i-th tire, respectively; fl, fr, rl, rr represent the left front tire, right front tire, left rear tire, and right rear tire, respectively; μ is the road adhesion coefficient; T i The torque r for each wheel w The tire's rolling radius;

[0049] The constraints are:

[0050]

[0051] |T i |≤|min(μ i F zi r w,T max )|,i=fl,fr,rl,rr

[0052] Among them, T max This indicates the maximum output capacity of the hub motor.

[0053] A distributed drive electric vehicle path tracking and stability control system includes:

[0054] State-space equation construction module: Based on the driver path tracking model and the vehicle three-degree-of-freedom motion model, construct the state-space equation;

[0055] Envelope region construction module: Establishes a dynamic envelope region based on the phase plane method, inputs the current vehicle parameters into the dynamic stable envelope region, and outputs the weights;

[0056] Control module: The controller is constructed based on the state-space equation, and the weights of the output of the dynamic stable envelope region are used as the control weights of the controller. Then, the vehicle state variables and vehicle parameters are input into the controller, and the outputs the longitudinal driving torque required for driving, the additional yaw torque, the additional front wheel steering angle, and the driver's intended steering angle. The additional front wheel steering angle is superimposed with the driver's intended steering angle as the input of the steering motor.

[0057] Torque Distribution Module: Design a torque distributor to distribute the longitudinal drive torque and additional yaw torque, and use the distributed torque as the setpoint for the four hub motors.

[0058] A computer storage medium storing a readable program that, when executed, instructs a computing device to perform the distributed drive electric vehicle path tracking and stability control method described above.

[0059] An electronic device includes: a processor, a memory, a communication interface, and a communication bus, wherein the processor, the memory, and the communication interface communicate with each other through the communication bus;

[0060] The memory is used to store at least one executable instruction, which causes the processor to perform operations corresponding to the distributed drive electric vehicle path tracking and stability control method described above.

[0061] A computer program product includes computer instructions, characterized in that the computer instructions instruct a computing device to perform operations corresponding to the distributed drive electric vehicle path tracking and stability control method described above.

[0062] The beneficial effects of this invention are:

[0063] 1. This invention establishes a state-space equation considering the vehicle's lateral and longitudinal deviations, yaw rate, and path prediction errors. The AFS agent uses the path prediction error as a cost function, while the DYC agent uses the centroid sideslip angle and path prediction error as cost functions, decoupling the control objectives. AFS focuses on reducing lateral deviation and heading error, while DYC focuses on suppressing centroid sideslip angle and yaw response. Their combined action ensures that the vehicle maintains good handling without instability under extreme conditions. Furthermore, through decoupling, the controller exhibits stronger robustness and scalability, facilitating dynamic weight adjustment under different operating conditions and improving the vehicle's overall control performance. In addition, this invention designs a vehicle stability envelope region based on phase plane theory and a fuzzy controller to coordinate the control weights of the objective function, which is beneficial for coordinating the control objective and controller output under complex conditions, thus improving control effectiveness.

[0064] 2. By establishing a driver path tracking model and a three-degree-of-freedom vehicle dynamics model, the coupling relationship between the driver's operating intention and the vehicle's motion state can be reflected more realistically, thereby improving the adaptability of the control algorithm to actual driving conditions and enhancing the accuracy and stability of vehicle trajectory tracking.

[0065] 3. By formalizing the vehicle state-space equations, this invention provides a unified mathematical description for the design of predictive controllers, making it easier to incorporate key state variables such as yaw rate, sideslip angle, lateral deviation, and heading deviation into the control framework, which is beneficial to improving the systematicness and scalability of controller design.

[0066] 4. By introducing a method for determining the dynamic stability envelope region, this invention can dynamically limit the stable operating boundary of a vehicle under different vehicle speeds and different road surface adhesion coefficients, thereby achieving real-time monitoring and early warning of the vehicle's operating status and ensuring that the vehicle still has good handling stability under extreme conditions.

[0067] 5. By designing the drive, active front wheel steering, and direct yaw torque as independent intelligent agents and setting corresponding cost functions, this invention can effectively coordinate the control objectives of different actuators, taking into account longitudinal speed maintenance, path tracking, and yaw stability, thereby achieving comprehensive optimization of multiple objectives and improving the flexibility and robustness of the control system.

[0068] 6. By designing a torque distributor and constructing an optimized cost function, this invention can reasonably distribute the driving torque of each wheel while satisfying the actuator constraints, avoid oversaturation of a single wheel, reduce the difference in tire adhesion utilization, thereby improving the driving efficiency and tire utilization of the whole vehicle, extending the actuator life and improving the energy efficiency and safety of the whole vehicle. Attached Figure Description

[0069] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, for those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0070] Figure 1 This is a schematic diagram of the control strategy framework of the present invention;

[0071] Figure 2 This is a schematic diagram of the driver pre-aiming model of the present invention;

[0072] Figure 3 This is a schematic diagram of the three-degree-of-freedom vehicle dynamics model of the present invention;

[0073] Figure 4 This is a diagram showing the variation of the stable envelope region under different road surface adhesion coefficients according to the present invention;

[0074] Figure 5 This is a diagram showing the variation of the stable envelope region under different vehicle longitudinal speeds according to the present invention;

[0075] Figure 6 This is a schematic diagram illustrating the calculation of the stability factor in this invention;

[0076] Figure 7 The relationship between the front wheel sideslip utilization rate, the rate of change of steering input, and the AFS penalty weight is derived from the fuzzy controller.

[0077] Figure 8 This is a schematic diagram comparing the vehicle path under high-speed, high-adhesion conditions with the ideal path.

[0078] Figure 9 This is a graph showing the change in the vehicle's center of gravity sideslip angle under high-speed, high-adhesion conditions.

[0079] Figure 10 This is a graph showing the front wheel steering angle of a vehicle under high-speed, high-adhesion conditions.

[0080] Figure 11 A graph showing the torque distribution of four-wheel drive under high-speed, high-adhesion conditions;

[0081] Figure 12 This is a schematic diagram comparing the vehicle path and the ideal path under medium-speed, low-adhesion conditions.

[0082] Figure 13 This is a graph showing the change in the vehicle's center of gravity sideslip angle under medium-speed, low-adhesion conditions.

[0083] Figure 14 This is a curve of the front wheel steering angle of a vehicle under medium-speed, low-adhesion conditions.

[0084] Figure 15This is a curve showing the torque distribution of four-wheel drive under medium-speed, low-adhesion conditions. Detailed Implementation

[0085] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0086] Example 1

[0087] This embodiment proposes a distributed drive vehicle path tracking and stability control method. It primarily establishes state-space equations using a driver path tracking model and a vehicle three-degree-of-freedom motion model. A distributed model prediction algorithm is then used to establish an integrated drive-braking, AFS (Adaptive Front-Side Path) and DYC (Dynamic Yard Control) path tracking and stability controller, outputting drive torque, additional front wheel steering angle, and yaw moment. The drive torque and yaw moment are fed into a torque distribution layer for four-wheel torque distribution; thereby controlling the vehicle to track the desired speed, follow the ideal path, and maintain vehicle stability.

[0088] Specifically, such as Figure 1 As shown, the distributed drive electric vehicle path tracking and stability control method includes the following steps:

[0089] S1, based on the driver path tracking model and the vehicle three-degree-of-freedom motion model, construct the state-space equation;

[0090] Driver path tracking model, such as Figure 2 As shown, the vehicle is considered to be traveling along the prescribed road centerline. The state variable characterizing the vehicle's tracking accuracy can be calculated from the vehicle's current state and the driver's aiming distance, and is expressed as:

[0091]

[0092] Among them, L p e represents the driver's aiming distance. y This represents the lateral deviation between the vehicle's current position and the near-target point. v represents the heading error between the vehicle's current heading and the reference direction. y and v x These represent the vehicle's lateral velocity and longitudinal velocity, respectively; γ represents the vehicle's yaw rate; ρ r Represents the curvature of the reference path.

[0093] A three-degree-of-freedom motion model of a vehicle considering longitudinal, lateral, and yaw motion is as follows: Figure 3 As shown, it can be represented as:

[0094]

[0095] Where m represents the total mass of the vehicle, v x β represents the vehicle's longitudinal velocity, γ represents the sideslip angle, γ represents the vehicle's yaw rate, and F represents the vehicle's longitudinal velocity. xfl F xfr F xrl F xrr F represents the longitudinal force acting on the left front wheel, right front wheel, left and right wheels, and right rear wheel, respectively. yfl F yfr F yrl F yrr F represents the lateral force F acting on the left front wheel, right front wheel, left and right wheels, and right rear wheel, respectively. yf F yr I represents the lateral forces acting on the front and rear wheels of the vehicle, respectively. z The moment of inertia of a vehicle, l f l r These represent the distances between the front and rear axles and the vehicle's center of gravity, respectively; δ represents the front wheel steering angle, input by the driver (the driver's intended steering angle). driver The additional front wheel steering angle δ calculated by the AFS system AFS The values ​​are obtained by superposition; ΔM represents the additional yaw torque. Considering that the steering angle of the front wheels is relatively small under vehicle steering conditions, we can assume cosδ≈1, sinδ≈0. We also assume that the tire lateral slip characteristics are within the linear range and ignore the influence of the wheelbase. The lateral forces on the left and right tires can be combined. Therefore, the total lateral force on the front and rear axles of the vehicle can be expressed as:

[0096] F yf =-C f α f F yr =-C r α r (3)

[0097] Among them, F yf and F yr C represents the lateral forces acting on the front and rear wheels of the vehicle, respectively. f and C r These represent the lateral stiffness of the front and rear axles of the vehicle, respectively; α f and α r These represent the slip angles of the front and rear wheels, respectively.

[0098] The expression for the wheel slip angle is as follows:

[0099] α f =β+l f r / v x -δαr =β-l r r / v x (4)

[0100] Where δ is the front wheel steering angle of the vehicle;

[0101] Substituting equations (3), (4), and (5) into (2) and combining them with (1), the resulting state-space equation is:

[0102]

[0103] Where, u = [F x ,δ AFS ,ΔM] T ,d=[δ driver ,ρ r ] T ,

[0104]

[0105] Where u is the control input, F x Let d be the longitudinal force required for wheel drive, and A be the system disturbance quantity. t Let B be the state transition matrix of the system. t Let E be the input transition matrix of the system. t Let X be the system's disturbance transfer matrix; X is the vehicle state variable.

[0106] S2, establish a dynamic stable envelope region based on the phase plane method, input the current vehicle parameters into the dynamic envelope region, and output the weights;

[0107] After establishing the dynamic model, a dynamic envelope stable region is established based on the phase plane method to coordinate the controller weights. Its inputs are the current state of the vehicle body, including: yaw rate, sideslip angle, longitudinal vehicle speed, road adhesion coefficient, and vehicle body parameters; the outputs are the control weights of the DMPC controller.

[0108] The envelope region is quadrilateral, with four vertices (r) C ,β C ),(r D ,β D ),(-r C ,-β C ),(-r D ,-β D The calculation formula for ) is as follows:

[0109]

[0110] Where, α sl,f Indicates the saturation value of the front wheel slip angle, α sl,rδ represents the saturation value of the rear wheel slip angle. max denoted as the maximum steering angle of the front wheel, μ as the road adhesion coefficient, and g as the acceleration due to gravity.

[0111] Figure 4 and Figure 5 These represent the stable envelope regions as a function of the road adhesion coefficient and longitudinal vehicle speed, respectively. Based on the established dynamic stable envelope regions, a stability requirement factor P is designed according to the vehicle's sideslip angle and yaw rate. stable It can be represented as:

[0112]

[0113] Among them, S AB (h),S BC (h),S CD (h),S AD (h) represents the distances between the stable point and the four boundaries of the envelope region, respectively. ρ1 and ρ2 are weighting factors; see [link to details]. Figure 6 This is used to measure the degree of risk near the boundary. It is worth noting that the purpose of determining the envelope region in this embodiment is to ensure that the supervisor can correctly identify and issue weight change signals to the controller when the vehicle state approaches the envelope boundary. Therefore, this embodiment uses g(x) obtained by translating and scaling the sigmoid function to characterize this sensing behavior. It is expressed as:

[0114]

[0115] When x = 0.5, g(x) = 0.1, while when x = 0.3, g(x) = 0.69. This indicates that the supervisor has good perception ability when the vehicle state is close to the boundary of the boundary envelope region, so that the vehicle state operates within the boundary of the envelope region, while it is not sensitive when it is close to the center of the region, thus avoiding overly conservative control.

[0116] The weighting factor in stable mode is expressed as:

[0117]

[0118] Where, λ DYC λ represents the weighting factor of the cost function of the yaw moment agent. AFS k is the weighting factor of the cost function of the front-wheel steering agent. s3 k is the coordination coefficient of the yaw moment agent. s2 q1 and q2 are the coordination coefficients of the front wheel steering agent, and are control parameters that are adjusted according to the actual simulation results.

[0119] Furthermore, the control limitations of AFS are reflected in the fact that when the tires enter the nonlinear region, further increasing the front wheel steering angle cannot increase the corresponding lateral force, and may even lead to further vehicle instability. Also, considering that if the AFS system forcibly intervenes during a driver's sudden obstacle avoidance maneuver, it may conflict with the driver's actual expectations, this embodiment uses a fuzzy controller to calculate the penalty weight η of AFS. AFS The controller inputs are the front wheel slip angle utilization rate Δα and the driver steering input change rate Δδ, where the slip angle utilization rate Δα is defined as the ratio of the front wheel slip angle to the saturation slip angle, expressed as:

[0120]

[0121] Wherein, the saturation side slip angle α max For the critical values ​​of the nonlinear and linear intervals, satisfying Analyze the effects of road surface adhesion coefficient and vertical load on the front wheel saturation sideslip angle. Due to the change in vertical load, α... max The impact of this on α is relatively small; therefore, to simplify calculations, this embodiment ignores its effect on α. max The influence of this. Therefore, under a fixed vertical load, the road adhesion coefficient and the saturation slip angle α can be fitted. max The relationship yields:

[0122] α max =f(μ)=8.75μ 2 -2.271μ+3.467 (11)

[0123] Design a fuzzy controller to calculate η AFS The fuzzy subset of Δα is {S, M, L, XL}, and the fuzzy subset of Δδ is {S, M, L, XL}, η AFS The fuzzy subset is {S, M, L, XL}, and the fuzzy control rules are shown in Table 1; Δα, Δδ, and η AFS The three-dimensional mapping relationship is as follows Figure 7 As shown;

[0124] Table 1 Fuzzy Control Rules

[0125]

[0126]

[0127] S3, a DMPC controller is constructed based on the state-space equation, and the weights of the output of the dynamic stable envelope region are used as the control weights of the controller; then, the vehicle state variables and vehicle parameters are input into the controller, and the outputs are the longitudinal driving torque required for driving, the additional yaw torque, the additional front wheel steering angle, and the driver's intended steering angle (driver steering angle input δ). driverThe additional front wheel steering angle will be superimposed on the driver's intended steering angle as the input to the steering motor;

[0128] The controller comprises three agents: a drive agent, a front-wheel steering agent, and a yaw torque agent. The longitudinal control system consists of only one drive agent, whose purpose is to calculate the total longitudinal driving force required to track the vehicle's longitudinal velocity. The lateral control system includes a front-wheel steering agent and a yaw torque agent, and its control objectives are path tracking and lateral stability.

[0129] When deriving the prediction state expression for each agent, the inputs of the other two agents are considered as a variable disturbance sequence d. i,1 (k), and simultaneously in the prediction time domain N p The driver is assumed to input steering angle δ driver curvature ρ of the reference path r Since it remains unchanged, it is considered as a fixed interference sequence d2(k).

[0130] The incremental system state equations are used for derivation. According to equation (5), the system state expressions of each agent are obtained after discretization:

[0131]

[0132] in:

[0133]

[0134] Where A is the state variable transition matrix, B i For the control variable transition matrix, E i,1 C is the neighbor agent information transfer matrix, E2 is the interference transfer matrix, and C is the interference transfer matrix. i E is the output transition matrix. i,1 E1 represents the coefficient matrix of the variable interference sequence for the i-th agent; E2 represents the coefficient matrix of the fixed interference sequence; B i C represents the control input matrix for the i-th agent; i y represents the output matrix of the i-th agent; i (k) and y i (k-1) represent the system output of the i-th agent at time t=k and the system output at the previous time, respectively; u i (k) and u i (k-1) represent the control quantities of the i-th agent at time t=k and t=k-1, respectively; d i (k) and d i (k-1) represents the disturbance amount of the i-th agent at times t=k and t=k-1, respectively; T represents the control period; I represents the identity matrix.

[0135] This embodiment uses the method of exchanging the historical best control sequences among the agents to construct the variable disturbance sequence. After removing the already implemented first term from the historical best control information of the neighboring agents at the previous time step, the last control variable is copied at the end of the control sequence to obtain the variable disturbance sequence of the i-th agent, thereby calculating the future state in the time domain. The purpose of copying the last control variable is to ensure that the control sequence dimensions are the same. This can be represented as:

[0136] D i,1 (k)=[d i,1 (k+1|k-1),d i,1 (k+2|k-1), ...,d i,1 (k+N c -1|k-1),Δd i,1 (k+1|k-1)] T (13)

[0137] Among them, D i,1 (k) After extracting the historical optimal control variables from the previous time step for the neighboring agents, the known disturbance information is obtained, N c The number of system control variables;

[0138] In the DMPC framework, the accuracy of the exchanged historical control sequences gradually decreases as the prediction step size increases. To mitigate this problem, this embodiment introduces an attenuation mechanism to weaken the impact of outdated control information, thereby improving the reliability of the coordination process. The variable interference sequence after incorporating the attenuation mechanism is shown below. for:

[0139]

[0140] Where α is the decay factor and n is the decay threshold step number.

[0141] Derivation of the prediction of y in the time domain i (k):

[0142] y i (k+1|k)=C i AΔx(k)+C i E i,1 Δd i,1 (k)+C i E2Δd2(k)+C i B i Δu i (k)+y i (k) (15)

[0143]

[0144] Among them, y i(k+i|k) represents the prediction output at step i when prediction is performed at time k; p represents the prediction time domain; m represents the control time domain;

[0145] For ease of representation, the following definition is provided:

[0146]

[0147] The derivation leads to:

[0148]

[0149] in:

[0150]

[0151] After obtaining the system output in the prediction time domain, the next step is to define the objective cost function for each agent:

[0152]

[0153] In the formula, J1, J2, and J3 represent the cost functions of the driving agent, the front wheel steering agent, and the yaw torque agent, respectively.

[0154] The control objective of the longitudinal drive agent is to track the desired vehicle speed; the control objective of the AFS agent is to reduce lateral and heading angle errors; and the control objective of the DYC agent is to suppress the vehicle's center of gravity sideslip angle as much as possible. The penalty weight matrix for the control increment aims to ensure smooth control. To avoid mutually canceling control behaviors between agents in the distributed MPC framework (e.g., AFS applies an additional rightward steering angle, while DYC simultaneously applies a leftward yaw moment), a penalty term for the actual control quantity is introduced into the cost function, serving as the terminal penalty weight.

[0155] Constraint settings: Based on the established envelope region, constraints are applied to the yaw rate and rear wheel sideslip angle. Simultaneously, constraints are also needed on the output capabilities of each actuator, including output increment constraints and control quantity constraints. These constraints are used to limit the additional front wheel steering angle and additional yaw moment, as well as their variations. These constraints can be expressed as:

[0156]

[0157] Where, N c Let u(k+j) and Δu(k+j) represent the input control value and input control increment at step j, respectively. max and u min These represent the maximum and minimum input values, Δu, respectively. max and Δu min Representing the maximum and minimum values ​​of the input increment, γ(k+i) and α, respectively.r (k+i) represent the yaw rate and the sideslip angle at step j, respectively, and γ max (k+i) and γ min (k+i) represent the maximum and minimum values ​​of the yaw rate, respectively, α rmax (k+i) and α rmin (k+i) represent the maximum and minimum values ​​of the centroid sideslip angle, respectively.

[0158] After calculating the control increment sequence at the current moment, the first term is added to the control quantity at the previous moment to obtain the longitudinal torque at the current moment, which is then used as the input to the lower-level torque distributor.

[0159] S4, Design a torque distributor to distribute the longitudinal drive torque and the additional yaw torque, and use the distributed torque as the setpoint for the four hub motors;

[0160] After the DMPC controller is designed, it is necessary to distribute the longitudinal drive torque and additional yaw torque it outputs. Therefore, a torque distributor needs to be designed to use the distributed torque as the setpoint for the four hub motors.

[0161] Torque distribution can be described as a convex quadratic programming (QP) problem.

[0162] Define the longitudinal attachment utilization rate λ i for:

[0163]

[0164] Among them, F xi F zi Let fl, fr, rl, rr represent the longitudinal force and vertical load of the i-th tire, respectively; fl, fr, rl, rr represent the left front tire, right front tire, left rear tire, and right rear tire, respectively; μ is the road adhesion coefficient; T i The torque for each wheel.

[0165] The cost function is constructed using the longitudinal adhesion utilization rate and utilization variance of the four tires:

[0166]

[0167] Among them, J torque The cost function of the torque distributor. The average longitudinal adhesion utilization rate; w1 and w2 are weighting factors; the longitudinal force F of the i-th tire. xi and T i The conversion relationship is as follows:

[0168] F xi r w =T i,i=fl,fr,rl,rr (25)

[0169] Where, r w The tire's rolling radius;

[0170] The constraints include yaw torque and longitudinal moment constraints, hub motor output torque constraints, and the requirement that the distributed torque cannot exceed the maximum value provided by the road surface. These constraints can be expressed as:

[0171]

[0172] |T i |≤|min(μ i F zi r w ,T max )|,i=fl,fr,rl,rr (27)

[0173] Among them, T max This indicates the maximum output capacity of the hub motor;

[0174] The distribution results are used as the given values ​​for the four hub motors to complete the torque distributor design.

[0175] Based on a similar inventive concept, embodiments of the present invention also provide a computer storage medium storing a readable program that, when run by a processor, can execute the aforementioned distributed drive electric vehicle path tracking and stability control method.

[0176] Based on a similar inventive concept, this invention provides an electronic device, including: a processor, a memory, a communication interface, and a communication bus, wherein the processor, the memory, and the communication interface communicate with each other through the communication bus;

[0177] The memory is used to store at least one executable instruction, which causes the processor to perform the operation corresponding to the above-described distributed drive electric vehicle path tracking and stability control method.

[0178] Based on a similar inventive concept, embodiments of the present invention also provide a computer program product, including computer instructions, which instruct a computing device to perform operations corresponding to the above-described distributed drive electric vehicle path tracking and stability control method.

[0179] Example 2

[0180] In this embodiment, the method proposed in Example 1 was verified by joint simulation using MATLAB / Simulink and CarSim;

[0181] 1) High-speed, high-adhesion working conditions

[0182] A double lane change test was conducted on the vehicle under the conditions of a longitudinal speed of 100 km / h and a road surface adhesion coefficient of 0.85. The experimental procedure was as follows:

[0183] This embodiment uses a co-simulation platform of MATLAB R2023b / Simulink and CarSim 2023.1: the processor is an Intel Core i7-12700F, and a pre-built .slx file is set in CarSim to build the co-simulation platform. The scenario is set as a high-speed, high-adhesion double lane change, with a longitudinal target vehicle speed of 100 km / h and a road adhesion coefficient μ = 0.85. CarSim provides high-fidelity models of the road and the vehicle / tire, and Simulink implements the controller. The co-simulation uses a fixed step size of 0.01s, with ode3 (Bogacki–Shampine) as the Simulink solver. Throughout the process, lane trajectory, four-wheel torque distribution results, and key parameters such as the center of gravity sideslip angle are synchronously saved via To Workspace for subsequent statistical and comparative analysis. Vehicle parameters are shown in Table 1 below.

[0184] Table 1 Vehicle Parameter Statistics Table

[0185]

[0186]

[0187] The verification results are as follows Figures 8-11 As shown: Among them, such as Figure 8 As shown, the vehicle's trajectory can follow the ideal path quite well; as Figure 9 and Figure 10 As shown, both the front wheel steering angle and the sideslip angle are kept within a reasonable range, ensuring the vehicle's handling stability; Figure 11 As shown, the torque distribution of the four-wheel drive is reasonable, and there is no phenomenon of single-wheel overload.

[0188] 2) Medium-speed, low-adhesion working conditions

[0189] The same double lane change test was conducted on the vehicle under the conditions of longitudinal speed of 70km / h and road surface adhesion coefficient of 0.5. The experimental procedure and hardware configuration were the same as those in Experiment 1, except that the longitudinal speed and road surface adhesion coefficient were changed.

[0190] The verification results are as follows Figures 12-15 As shown: Among them, such as Figure 12 As shown, the vehicle path still maintains a high degree of consistency with the ideal path; for example... Figure 13 and Figure 14 As shown, the front wheel steering angle and center of gravity sideslip angle remained within the allowable range, and no instability occurred; Figure 15As shown, the torque distribution of the four-wheel drive is balanced, and the longitudinal adhesion utilization rate of the tires remains within a reasonable range; this indicates that the method can also ensure the yaw stability and path tracking performance of the vehicle under low adhesion conditions.

[0191] Example 3

[0192] This embodiment proposes a distributed drive electric vehicle path tracking and stability control system, specifically including:

[0193] State-space equation construction module: Based on the driver path tracking model and the vehicle three-degree-of-freedom motion model, construct the state-space equation;

[0194] Envelope region construction module: Establishes a dynamic envelope region based on the phase plane method, inputs the current vehicle parameters into the dynamic stable envelope region, and outputs the weights;

[0195] Control module: The controller is constructed based on the state-space equation, and the weights of the output of the dynamic stable envelope region are used as the control weights of the controller. Then, the vehicle state variables and vehicle parameters are input into the controller, and the outputs the longitudinal driving torque required for driving, the additional yaw torque, the additional front wheel steering angle, and the driver's intended steering angle. The additional front wheel steering angle is superimposed with the driver's intended steering angle as the input of the steering motor.

[0196] Torque distribution module: Design a torque distributor to distribute the longitudinal drive torque and additional yaw torque, and use the distributed torque as the setpoint for the four hub motors;

[0197] The methods of the present invention can be implemented in hardware, firmware, or as software or computer code that can be stored in a recording medium (such as a CD-ROM, RAM, floppy disk, hard disk, or magneto-optical disk), or as computer code originally stored on a remote recording medium or a non-transitory machine-readable medium and subsequently stored on a local recording medium, downloaded via a network. Thus, the methods described herein can be processed by software stored on a recording medium using a general-purpose computer, a dedicated processor, or programmable or dedicated hardware (such as an ASIC or FPGA). It is understood that the computer, processor, microprocessor controller, or programmable hardware includes storage components (e.g., RAM, ROM, flash memory, etc.) capable of storing or receiving software or computer code that, when accessed and executed by the computer, processor, or hardware, implements the methods described herein. Furthermore, when a general-purpose computer accesses the code used to implement the methods shown herein, the execution of the code transforms the general-purpose computer into a dedicated computer for performing the methods shown herein.

[0198] The foregoing has shown and described the basic principles, main features, and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The embodiments and descriptions in the specification are merely illustrative of the principles of the invention. Various changes and modifications can be made to the invention without departing from its spirit and scope, and all such changes and modifications fall within the scope of the claimed invention.

Claims

1. A method for path tracking and stability control of a distributed-drive electric vehicle, characterized in that, Includes the following steps: Based on the driver path tracking model and the vehicle three-degree-of-freedom motion model, a state-space equation is constructed. A dynamic envelope region is established based on the phase plane method, and the current vehicle parameters are input into the dynamic stable envelope region, and the weights are output. The controller is constructed based on the state-space equation, and the weights of the output of the dynamic stable envelope region are used as the control weights of the controller. Then, the vehicle state variables and vehicle parameters are input into the controller, which outputs the longitudinal driving torque required for driving, the additional yaw torque, the additional front wheel steering angle, and the driver's intended steering angle; the additional front wheel steering angle will be superimposed with the driver's intended steering angle as the input to the steering motor; Design a torque distributor to distribute the longitudinal drive torque and additional yaw torque, and use the distributed torque as the setpoint for the four hub motors.

2. The distributed drive electric vehicle path tracking and stability control method according to claim 1, characterized in that, The driver path tracking model is as follows: Among them, L p e represents the driver's aiming distance. y This represents the lateral deviation between the vehicle's current position and the near-target point. v represents the heading error between the vehicle's current heading and the reference direction. y and v x These represent the vehicle's lateral velocity and longitudinal velocity, respectively; γ represents the vehicle's yaw rate; ρ r The curvature representing the reference path; The three-degree-of-freedom motion model of the vehicle is as follows: F yf =-C f a f ,F yr =-C r a r a f =β+l f r / v x -da r =β-l r r / v x Where m represents the total mass of the vehicle, v x β represents the vehicle's longitudinal velocity, r represents the vehicle's sideslip angle, and F represents the vehicle's yaw rate. xfl F xfr F xrl F xrr F represents the longitudinal force acting on the left front wheel, right front wheel, left and right wheels, and right rear wheel, respectively. yfl F yfr F yrl F yrr F represents the lateral force F acting on the left front wheel, right front wheel, left and right wheels, and right rear wheel, respectively. yf F yr C represents the lateral forces acting on the front and rear axles of the vehicle, respectively. f and C r These represent the lateral stiffness of the front and rear axles of the vehicle, respectively; I z The moment of inertia of a vehicle, l f l r These represent the distances between the front and rear axles and the vehicle's center of gravity, respectively; δ represents the front wheel steering angle, determined by the driver's intended steering angle δ. driver The additional front wheel steering angle δ calculated by the AFS system AFS The result is obtained by superposition; ΔM represents the additional yaw torque; α f and α r These represent the slip angles of the front and rear wheels, respectively.

3. The distributed drive electric vehicle path tracking and stability control method according to claim 2, characterized in that, The state-space equation is: where, u = [F x , δ AFS , ΔM] T , d = [δ driver , ρ r T , ​ Where u is the control input, F x Let d be the longitudinal force required for wheel drive, and A be the system disturbance quantity. t Let B be the state transition matrix of the system. t Let E be the input transition matrix of the system. t Let X be the system's disturbance transfer matrix; X is the vehicle state variable.

4. The distributed drive electric vehicle path tracking and stability control method according to claim 3, characterized in that, The envelope region is quadrilateral, with four vertices (r) C ,β C ),(r D ,β D ),(-r C ,-β C ),(-r D ,-β D The calculation formula for ) is as follows: Where, α sl,f Indicates the saturation value of the front wheel slip angle, α sl,r δ represents the saturation value of the rear wheel slip angle. max denoted as the maximum steering angle of the front wheel, μ as the road adhesion coefficient, and g as the acceleration due to gravity.

5. The distributed drive electric vehicle path tracking and stability control method according to claim 4, characterized in that, The controller includes: a drive agent, a front wheel steering agent, and a yaw torque agent; the drive agent constitutes a longitudinal control system to calculate the total longitudinal driving force required to track the longitudinal speed of the vehicle; the front wheel steering agent and the yaw torque agent constitute a lateral control system, with the control objectives being path tracking and lateral stability. The objective cost functions for each agent are as follows: In the formula, J1, J2, and J3 represent the cost functions of the driving agent, the front wheel steering agent, and the yaw torque agent, respectively. The constraints include: you min ≤u(k+j)≤u max Δu min ≤Δu(k+j)≤Δu max γ min (k+i)≤γ(k+i)≤γ max (k+i) α rmin (k+i)≤α r (k+i)≤α rmax (k+i) Where, N c Let u(k+j) and Δu(k+j) represent the input control value and input control increment at step k, respectively. max and u min These represent the maximum and minimum input values, Δu, respectively. max and Δu min Representing the maximum and minimum values ​​of the input increment, γ(k+i) and α, respectively. r (k+i) represent the yaw rate and the sideslip angle at the i-th step, respectively, γ max (k+i) and γ min (k+i) represent the maximum and minimum values ​​of the yaw rate, respectively, α rmax (k+i) and α rmin (k+i) represent the maximum and minimum values ​​of the centroid sideslip angle, respectively.

6. The distributed drive electric vehicle path tracking and stability control method according to claim 5, characterized in that, The cost function of the torque distributor is: F xi r w =T i ,i=fl,fr,rl,rr Among them, J torque The cost function of the torque distributor. λ represents the average longitudinal attachment utilization rate; w1 and w2 are weighting factors; λ i To define the longitudinal attachment utilization rate, F xi F zi Let fl, fr, rl, rr represent the longitudinal force and vertical load of the i-th tire, respectively; fl, fr, rl, rr represent the left front tire, right front tire, left rear tire, and right rear tire, respectively; μ is the road adhesion coefficient; T i The torque r for each wheel w The tire's rolling radius; The constraints are: Among them, T max This indicates the maximum output capacity of the hub motor.

7. A distributed drive electric vehicle path tracking and stability control system, characterized in that, include: State-space equation construction module: Based on the driver path tracking model and the vehicle three-degree-of-freedom motion model, a state-space equation is constructed. Envelope region construction module: Establishes a dynamic envelope region based on the phase plane method, inputs the current vehicle parameters into the dynamic stable envelope region, and outputs the weights; Control module: The controller is constructed based on the state-space equation, and the weights of the output of the dynamic stable envelope region are used as the control weights of the controller. Then, the vehicle state variables and vehicle parameters are input into the controller, and the outputs the longitudinal driving torque required for driving, the additional yaw torque, the additional front wheel steering angle, and the driver's intended steering angle. The additional front wheel steering angle is superimposed with the driver's intended steering angle as the input of the steering motor. Torque Distribution Module: Design a torque distributor to distribute the longitudinal drive torque and additional yaw torque, and use the distributed torque as the setpoint for the four hub motors.

8. A computer storage medium storing a readable program, characterized in that, When the program runs, it can instruct the computing device to execute the distributed drive electric vehicle path tracking and stability control method as described in any one of claims 1-6.

9. An electronic device, characterized in that, include: The processor, memory, communication interface, and communication bus are provided, wherein the processor, memory, and communication interface communicate with each other via the communication bus. The memory is used to store at least one executable instruction, which causes the processor to perform the operation corresponding to the distributed drive electric vehicle path tracking and stability control method as described in any one of claims 1-6.

10. A computer program product comprising computer instructions, characterized in that, The computer instructions instruct the computing device to perform the operations corresponding to the distributed drive electric vehicle path tracking and stability control method as described in any one of claims 1-6.

Images