A distributed steer-by-wire corner planning system and method

Abstract

The application discloses a kind of distributed steer-by-wire steering corner planning system and method, comprising: based on the current position of vehicle, obtain vehicle global reference information from map and road parameter database;Long-term vehicle dynamics planning model is constructed, with energy consumption minimization, vehicle flexibility enhancement and lateral stability improvement as optimization goal, to obtain the reference sequence of vehicle overall lateral force and yaw moment in long-term;Short-term distributed steer-by-wire steering system corner planning model is constructed, to obtain four-wheel corner reference target satisfying system constraint;According to four-wheel corner reference target satisfying system constraint, drive four steering actuators of vehicle to produce corresponding actual corner, to realize vehicle steering control.In the application, cloud side focuses on global optimal planning, and vehicle side focuses on real-time feasible planning, and both form closed-loop collaborative planning mechanism through two-way information interaction, which can significantly improve the multi-wheel coordination, driving stability and economy of distributed steer-by-wire steering vehicle.

Description

Technical Field

[0001] This invention belongs to the field of distributed steer-by-wire technology, specifically relating to a distributed steer-by-wire angle planning system and method. Background Technology

[0002] With the rapid development of vehicle intelligence and steer-by-wire technology, distributed steer-by-wire systems, with their advantages of flexible structure, rapid response, and independent four-wheel steering angle control, can achieve multiple steering modes such as same-direction, opposite-direction, and crabbing, giving vehicles greater maneuverability in complex road environments. This has made them the ideal steering system for intelligent electric vehicles. However, multi-wheel independent control also significantly increases the system's complexity and control difficulty. Achieving efficient and coordinated steering angle planning and control while simultaneously ensuring driving stability, vehicle energy consumption, and path tracking accuracy has become a key technical challenge that distributed steer-by-wire systems urgently need to overcome.

[0003] While some progress has been made in the research of distributed steer-by-wire systems in the prior art, there are still significant shortcomings. For example, Chinese invention patent application No. CN202010633638.4, entitled "A Control Method for Distributed Steering Drive System Based on Reinforcement Learning," achieves autonomous learning and improvement of independent steering angles and drive distribution for four wheels, comprehensively considering the vehicle's handling stability and tire durability, thus avoiding the limitations of traditional control methods. However, it suffers from slow training convergence speed and strong data dependence, making it difficult to guarantee the vehicle's overall performance under complex working conditions, and lacks forward-looking steering angle planning capabilities for multi-wheel collaboration. Chinese invention patent application No. CN202510095458.8, entitled "A Distributed Steer-by-Wire Vehicle Stability Control Method," improves the vehicle's yaw stability under uncertain conditions through robust control methods, but its control objective is limited to stability adjustment and fails to achieve global planning and coordination of distributed steer-by-wire.

[0004] With the development of vehicle-road-cloud integrated technology, vehicles can obtain global environmental information through the cloud, providing new technical conditions for long-term cornering planning. However, most existing cornering planning methods for distributed steer-by-wire systems are still limited to onboard local perception and short-term prediction, failing to effectively utilize the global perspective of the cloud for forward-looking, multi-time-domain collaborative planning. Summary of the Invention

[0005] In view of the shortcomings of the prior art, the purpose of this invention is to provide a distributed steer-by-wire angle planning system and method to solve the problem that the existing technology mainly relies on local information on the vehicle end for angle planning, and fails to make full use of the global environmental information provided by the cloud for forward-looking, multi-time-domain collaborative planning, which leads to insufficient coordination of multi-wheel angles, decreased steering stability and high vehicle energy consumption, making it difficult to achieve global optimization.

[0006] To achieve the above objectives, the technical solution adopted by the present invention is as follows:

[0007] The present invention provides a distributed steer-by-wire angle planning system, comprising: a cloud information acquisition unit, a cloud long-term domain planning unit, a vehicle-cloud information interaction unit, a vehicle-end information acquisition unit, a vehicle-end short-term domain planning unit, and a steering actuator control unit;

[0008] The cloud information acquisition unit is used to acquire vehicle global reference information and send the vehicle global reference information to the cloud long-term domain planning unit and the vehicle-cloud information interaction unit.

[0009] The cloud-based long-term planning unit is used to receive the vehicle global reference information obtained by the cloud-based information acquisition unit and the vehicle real-time status information sent by the vehicle-cloud information interaction unit, so as to construct a long-term vehicle dynamics planning model, solve for the reference sequence of the overall lateral force and yaw moment of the vehicle in the long-term domain, and send the reference sequence to the vehicle-cloud information interaction unit.

[0010] The vehicle-cloud information interaction unit is used to receive real-time vehicle status information collected by the vehicle-side information acquisition unit and send it to the cloud-based long-term domain planning unit, receive the overall vehicle lateral force and yaw moment reference sequence sent by the cloud-based long-term domain planning unit and the vehicle global reference information sent by the cloud-based information acquisition unit, and send them together to the vehicle-side short-term domain planning unit.

[0011] The vehicle-side information acquisition unit is used to collect real-time vehicle status information and send the real-time vehicle status information to the vehicle-cloud information interaction unit and the vehicle-side short-time domain planning unit.

[0012] The vehicle-side short-time domain planning unit is used to receive the overall lateral force and yaw moment reference sequence of the vehicle sent by the vehicle-cloud information interaction unit and the real-time vehicle status information sent by the vehicle-side information acquisition unit, construct a short-time domain distributed steer-by-wire system angle planning model, solve for the four-wheel angle reference target that meets the constraints, and send it to the steering actuator control unit.

[0013] The steering actuator control unit is used to control the vehicle's steering actuator to generate the corresponding actual steering angle based on the four-wheel steering angle reference target sent by the vehicle-side short-time domain planning unit, so as to achieve vehicle steering control.

[0014] Furthermore, the vehicle global reference information includes: future driving reference trajectory, future reference vehicle speed, road curvature, and road surface adhesion coefficient.

[0015] Furthermore, the real-time vehicle status information includes: longitudinal speed, yaw rate, sideslip angle, heading angle, and vehicle position.

[0016] This invention also provides a distributed steer-by-wire angle planning method, based on the above system, comprising the following steps:

[0017] 1) Based on the vehicle's current location, obtain global reference information for the vehicle from the map and road parameter database;

[0018] 2) Using the vehicle's global reference information and real-time vehicle status information as reference targets and initial states, a long-term vehicle dynamics planning model is constructed. With the optimization objectives of minimizing energy consumption, enhancing vehicle flexibility, and improving lateral stability, the reference sequence of the vehicle's overall lateral force and yaw moment in the long-term domain is obtained by solving the model.

[0019] 3) Based on the reference sequence of the overall lateral force and yaw moment of the vehicle in the long time domain and the real-time state information of the vehicle, a short-time domain distributed steer-by-wire system angle planning model is constructed, and the four-wheel steering angle reference target that satisfies the system constraints is obtained by solving the model.

[0020] 4) Based on the four-wheel steering angle reference target that satisfies the system constraints, drive the four steering actuators of the vehicle to generate the corresponding actual steering angles in order to achieve vehicle steering control.

[0021] Furthermore, step 2) is specifically as follows:

[0022] 21) Construct a long-term vehicle dynamics planning model;

[0023] Define the state vector of the long-time domain vehicle dynamics programming model for:

[0024] ;

[0025] In the formula, Sampling time, Sampling time The vehicle's center of gravity sideslip angle, Sampling time The vehicle's yaw rate, Sampling time The vehicle's heading angle, Sampling time The lateral position of the vehicle;

[0026] Input vector of long-term vehicle dynamics programming model The expressions for the overall lateral force and yaw moment of the vehicle are as follows:

[0027] ;

[0028] In the formula, Sampling time The overall lateral force of the vehicle, Sampling time The overall yaw moment of the vehicle;

[0029] Based on the lateral and yaw motion characteristics of the vehicle, a long-time-domain vehicle dynamics planning model is constructed as follows:

[0030] ;

[0031] in, The nonlinear function of the long-time-domain vehicle dynamics planning model is detailed below:

[0032] ;

[0033] In the formula, For vehicle quality, Let be the vehicle's moment of inertia about its vertical axis. Sampling time The longitudinal speed of the vehicle, The sampling period;

[0034] 22) Construct the overall objective function for long-term programming;

[0035] The equivalent steering resistance function of the vehicle is established as follows:

[0036] ;

[0037] In the formula, Sampling time The vehicle's equivalent steering resistance, and These are the front axle lateral stiffness and the rear axle lateral stiffness, respectively. and These are the front wheelbase and the rear wheelbase, respectively. For road curvature;

[0038] Based on the vehicle's equivalent steering resistance function, a long-term time-domain energy consumption objective function is constructed. for:

[0039] ;

[0040] In the formula, For the prediction time domain of the long-term vehicle dynamics planning model, For the first Predict the timing, and ; For the first Predict the longitudinal velocity of the vehicle at the next step. For the first The vehicle's equivalent steering resistance at the predicted moment;

[0041] Constructing a long-term trajectory tracking objective function as follows:

[0042] ;

[0043] In the formula, For the first The lateral position of the vehicle at the predicted moment. For the first The vehicle's reference lateral position at the predicted time step;

[0044] Using the vehicle's center of gravity sideslip angle as a key stability indicator, a vehicle stability objective function is constructed. as follows:

[0045] ;

[0046] In the formula, For the first The vehicle's sideslip angle at the predicted moment;

[0047] Combining the three objective functions of energy consumption, trajectory tracking, and stability, the overall objective function for long-term time-domain programming is constructed as follows:

[0048] ;

[0049] In the formula, Sampling time The overall objective function value of long-term programming. , , Weighting coefficients for different optimization objectives are used to achieve a balance between minimum energy consumption, trajectory accuracy, and attitude stability.

[0050] 23) Solve the long-term vehicle dynamics programming problem;

[0051] The reference sequence of overall lateral forces and yaw moments of the vehicle in the long time domain is solved using dynamic programming, as follows:

[0052] Based on the Bellman optimality principle, the sampling time is constructed. The optimal cost function of the overall objective function of time-domain programming Its expression is as follows:

[0053] ;

[0054] In the formula, Sampling time The optimal cost function value, It is a feasible set of the overall lateral forces and yaw moments of the vehicle;

[0055] according to The optimal cost function at time step, from stage The process proceeds recursively forward until the optimal cost function for all stages is obtained; this process involves enumerating the feasible set of overall vehicle lateral forces and yaw moments. And by combining the long-term vehicle dynamics planning model, the system state of the next stage can be obtained, thus achieving global search;

[0056] Constructing the optimal policy function Its expression is as follows:

[0057] ;

[0058] In the formula, Sampling time The optimal policy function value;

[0059] Calculations based on long-time domain vehicle dynamics planning model State vector at time step ,as follows:

[0060] ;

[0061] Finally, a reference sequence of the overall lateral force and yaw moment of the vehicle in the long-term domain is obtained, expressed as follows:

[0062] ;

[0063] in, For the first The overall lateral force of the reference vehicle at the predicted step time. For the first The reference vehicle yaw moment at the predicted step.

[0064] Furthermore, step 3) is specifically as follows:

[0065] 31) Construct a short-time-domain distributed steer-by-wire system angle planning model;

[0066] Define the state vector of the steering angle planning model for a short-time distributed steering-by-wire system. ,as follows:

[0067] ;

[0068] Input vector The four wheel corners are selected as follows:

[0069] ;

[0070] In the formula, Sampling time The turning angles of the left front wheel, right front wheel, left rear wheel, and right rear wheel;

[0071] The following is a model for the steering angle planning of a short-time distributed steer-by-wire system:

[0072] ;

[0073] In the formula, for The state vector of the short-time-domain distributed steering-by-wire system's angle planning model. The nonlinear function of the steering angle planning model for a short-time domain distributed steer-by-wire system is detailed below:

[0074] ;

[0075] In the formula, , , , , Sampling time The lateral forces of the left front tire, the right front tire, the left rear tire, and the right rear tire;

[0076] The lateral forces of the tire are described using a linear tire model, as follows:

[0077] ;

[0078] In the formula, , , , These are the lateral stiffness of the left front tire, the right front tire, the left rear tire, and the right rear tire, respectively. , , , Sampling time The slip angles of the left front tire, right front tire, left rear tire, and right rear tire are calculated using the following formulas:

[0079] ;

[0080] 32) Construct the overall objective function for short-time domain programming;

[0081] Constructing a short-time-domain trajectory tracking objective function as follows:

[0082] ;

[0083] In the formula, For the prediction time domain of the steering angle planning model of a short-time distributed steer-by-wire system, For the first Predict the timing, and , For the first Predict the lateral position of the vehicle at any given moment. For the first The reference lateral position of the vehicle at the predicted moment;

[0084] The overall lateral force and yaw moment of the vehicle can be calculated from the lateral forces of the four wheels of the vehicle, as follows:

[0085] ;

[0086] The tracking objective function of the overall lateral force and yaw moment reference sequence of the vehicle for:

[0087] ;

[0088] In the formula, For the first The overall lateral force of the vehicle at the predicted moment. For the first The overall lateral force of the reference vehicle at the predicted step time. For the first Predict the vehicle yaw moment at the next step. For the first Reference vehicle yaw moment at the predicted step;

[0089] Using the corner consistency objective function To improve the synchronization of the distributed steering-by-wire system, the objective function for steering angle consistency is... The expression is as follows:

[0090] ;

[0091] In the formula, For the first The left front wheel steering angle at the predicted moment. For the first The right front wheel steering angle at the predicted moment. For the first The left rear wheel steering angle at the predicted moment. For the first The right rear wheel steering angle at the predicted moment;

[0092] In summary, the overall objective function of short-time domain programming is... It can be written as:

[0093] ;

[0094] In the formula, , , The weighting coefficients are used to correspond to different optimization objectives, in order to achieve a comprehensive balance between trajectory tracking accuracy, cloud reference tracking performance, and corner consistency;

[0095] 33) Construct the steering angle planning constraints for the short-time domain distributed steering-by-wire system;

[0096] Based on the overall objective function of short-time domain programming, input constraints and state constraints are established for the steering angle planning model of a short-time domain distributed steering-by-wire system.

[0097] Set the maximum allowable range of steering angle for the distributed steering-by-wire system as follows:

[0098] ;

[0099] In the formula, For the first Predict the turning angle of each wheel at any given moment. and These are the minimum and maximum allowable turning angles for each wheel, respectively.

[0100] Constraints are imposed on the rate of change of the steering angle to prevent excessively rapid changes in the angle from causing control jitter, as follows:

[0101] ;

[0102] In the formula, This represents the maximum angle change within a single control cycle.

[0103] Constraints are imposed on the sideslip angle and yaw rate to ensure the vehicle maintains good handling stability, as follows:

[0104] ;

[0105] In the formula, For the first The vehicle's sideslip angle at the predicted moment. For the first The vehicle yaw rate at the predicted moment. and These are the maximum permissible values ​​for the sideslip angle and the maximum permissible value for the yaw rate, respectively.

[0106] 34) Solve the angle planning problem for a short-time distributed steering-by-wire system;

[0107] The short-time-domain distributed steer-by-wire system cornering planning model in step 31), the short-time-domain overall objective function in step 32), and the short-time-domain distributed steer-by-wire system cornering planning constraints in step 33) are combined into a finite-time-domain programming problem, which takes the following form:

[0108] ;

[0109] In the formula, Sampling time The predicted four-wheel turn sequence in the time domain is as follows:

[0110] ;

[0111] In the formula, For the first The input vector of the short-time-domain distributed steering-by-wire system angle planning model at the step prediction time;

[0112] The above finite-time programming problem can be represented as a constrained quadratic programming problem. The optimal control sequence can be obtained online using a fast quadratic programming solver or a sequential quadratic programming algorithm, as follows:

[0113] ;

[0114] in, The optimal control sequence obtained by solving the finite-time programming problem. For the first The optimal input vector corresponding to the prediction time step;

[0115] Through the above solution process, the four-wheel steering angle reference target sequence that satisfies the system constraints at the current sampling time is obtained. .

[0116] Further, step 4) specifically includes: collecting the actual turning angle of each wheel, constructing a turning angle closed-loop controller, inputting the error between the actual turning angle and the reference turning angle into the PID controller for adjustment, the PID controller calculates the drive control quantity of the steering motor according to the preset proportional term, integral term and derivative term, and drives each steering motor to output the corresponding current, so that the actual turning angle of the wheel gradually approaches the reference turning angle, realizing the tracking control of the four-wheel turning angle, ensuring that the vehicle can accurately track the reference trajectory and achieve optimal global performance.

[0117] The beneficial effects of this invention are:

[0118] In this invention, the cloud focuses on global optimal planning, while the vehicle focuses on real-time feasible planning. The two form a closed-loop collaborative planning mechanism through two-way information interaction, which can significantly improve the multi-wheel coordination, driving stability and economy of distributed steer-by-wire vehicles.

[0119] 1. This invention utilizes high-performance computing resources in the cloud and global environmental information to construct a long-term vehicle dynamics planning model, defines an equivalent steering resistance function, and uses the overall lateral force and yaw moment of the vehicle as planning control variables to achieve coordinated optimization of energy consumption, stability, and path tracking performance. Compared with traditional planning methods that rely on local perception information, this invention can predict future attitude changes of the vehicle in advance and generate a globally optimal reference target. This not only effectively reduces tire side slip loss and the frequency of steering angle correction, but also improves trajectory tracking accuracy, driving stability, and vehicle economy in complex road and high-speed scenarios.

[0120] 2. This invention constructs a short-time-domain distributed steer-by-wire system angle planning model on the vehicle side based on the vehicle's real-time status. This model can quickly track the reference sequence sent from the cloud and make corrections according to actual operating conditions, thereby ensuring the feasibility and robustness of the angle planning results in dynamic environments. Simultaneously, a consistency objective function is introduced on the vehicle side to effectively suppress angle oscillations caused by differences in wheel execution, enabling the vehicle to have higher stability and path-keeping capabilities under conditions such as high-speed lane changes, cornering, and low adhesion coefficients. Attached Figure Description

[0121] Figure 1 This is a schematic diagram of the system of the present invention.

[0122] Figure 2 This is a flowchart illustrating the method of the present invention. Detailed Implementation

[0123] To facilitate understanding by those skilled in the art, the present invention will be further described below with reference to embodiments and accompanying drawings. The content mentioned in the embodiments is not intended to limit the present invention.

[0124] Reference Figure 1 As shown, a distributed steer-by-wire angle planning system of the present invention includes: a cloud information acquisition unit, a cloud long-term domain planning unit, a vehicle-cloud information interaction unit, a vehicle-end information acquisition unit, a vehicle-end short-term domain planning unit, and a steering actuator control unit;

[0125] The cloud information acquisition unit is used to acquire vehicle global reference information and send the vehicle global reference information to the cloud long-term domain planning unit and the vehicle-cloud information interaction unit.

[0126] The cloud-based long-term planning unit is used to receive the vehicle global reference information obtained by the cloud-based information acquisition unit and the vehicle real-time status information sent by the vehicle-cloud information interaction unit, so as to construct a long-term vehicle dynamics planning model, solve for the reference sequence of the overall lateral force and yaw moment of the vehicle in the long-term domain, and send the reference sequence to the vehicle-cloud information interaction unit.

[0127] The vehicle-cloud information interaction unit is used to receive real-time vehicle status information collected by the vehicle-side information acquisition unit and send it to the cloud-based long-term domain planning unit, receive the overall vehicle lateral force and yaw moment reference sequence sent by the cloud-based long-term domain planning unit and the vehicle global reference information sent by the cloud-based information acquisition unit, and send them together to the vehicle-side short-term domain planning unit.

[0128] The vehicle-side information acquisition unit is used to collect real-time vehicle status information and send the real-time vehicle status information to the vehicle-cloud information interaction unit and the vehicle-side short-time domain planning unit.

[0129] The vehicle-side short-time domain planning unit is used to receive the overall lateral force and yaw moment reference sequence of the vehicle sent by the vehicle-cloud information interaction unit and the real-time vehicle status information sent by the vehicle-side information acquisition unit, construct a short-time domain distributed steer-by-wire system angle planning model, solve for the four-wheel angle reference target that meets the constraints, and send it to the steering actuator control unit.

[0130] The steering actuator control unit is used to control the vehicle's steering actuator to generate the corresponding actual steering angle based on the four-wheel steering angle reference target sent by the vehicle-side short-time domain planning unit, so as to achieve vehicle steering control.

[0131] The vehicle global reference information includes: future driving reference trajectory, future reference vehicle speed, road curvature, and road surface adhesion coefficient.

[0132] The real-time vehicle status information includes: longitudinal speed, yaw rate, sideslip angle, heading angle, and vehicle position.

[0133] Reference Figure 2 As shown, the present invention also provides a distributed steer-by-wire angle planning method, based on the above system, comprising the following steps:

[0134] 1) Based on the vehicle's current location, obtain global reference information for the vehicle from the map and road parameter database;

[0135] The vehicle global reference information includes: future driving reference trajectory, future reference vehicle speed, road curvature, and road surface adhesion coefficient.

[0136] The real-time vehicle status information includes: longitudinal speed, yaw rate, sideslip angle, heading angle, and vehicle position.

[0137] 2) Using the vehicle's global reference information and real-time vehicle status information as reference targets and initial states, a long-term vehicle dynamics planning model is constructed. With the optimization objectives of minimizing energy consumption, enhancing vehicle flexibility, and improving lateral stability, the reference sequence of the vehicle's overall lateral force and yaw moment in the long-term domain is obtained by solving the model.

[0138] Step 2) is as follows:

[0139] 21) Construct a long-term vehicle dynamics planning model;

[0140] Define the state vector of the long-time domain vehicle dynamics programming model for:

[0141] ;

[0142] In the formula, Sampling time, Sampling time The vehicle's center of gravity sideslip angle, Sampling time The vehicle's yaw rate, Sampling time The vehicle's heading angle, Sampling time The lateral position of the vehicle;

[0143] Input vector of long-term vehicle dynamics programming model The expressions for the overall lateral force and yaw moment of the vehicle are as follows:

[0144] ;

[0145] In the formula, Sampling time The overall lateral force of the vehicle, Sampling time The overall yaw moment of the vehicle;

[0146] Based on the lateral and yaw motion characteristics of the vehicle, a long-time-domain vehicle dynamics planning model is constructed as follows:

[0147] ;

[0148] in, The nonlinear function of the long-time-domain vehicle dynamics planning model is detailed below:

[0149] ;

[0150] In the formula, For vehicle quality, Let be the vehicle's moment of inertia about its vertical axis. Sampling time The longitudinal speed of the vehicle, The sampling period;

[0151] 22) Construct the overall objective function for long-term programming;

[0152] The equivalent steering resistance function of the vehicle is established as follows:

[0153] ;

[0154] In the formula, Sampling time The vehicle's equivalent steering resistance, and These are the front axle lateral stiffness and the rear axle lateral stiffness, respectively. and These are the front wheelbase and the rear wheelbase, respectively. For road curvature;

[0155] Based on the vehicle's equivalent steering resistance function, a long-term time-domain energy consumption objective function is constructed. for:

[0156] ;

[0157] In the formula, For the prediction time domain of the long-term vehicle dynamics planning model, For the first Predict the timing, and ; For the first Predict the longitudinal velocity of the vehicle at the next step. For the first The vehicle's equivalent steering resistance at the predicted moment;

[0158] Constructing a long-term trajectory tracking objective function as follows:

[0159] ;

[0160] In the formula, For the first The lateral position of the vehicle at the predicted moment. For the first The vehicle's reference lateral position at the predicted time step;

[0161] Using the vehicle's center of gravity sideslip angle as a key stability indicator, a vehicle stability objective function is constructed. as follows:

[0162] ;

[0163] In the formula, For the first The vehicle's sideslip angle at the predicted moment;

[0164] Combining the three objective functions of energy consumption, trajectory tracking, and stability, the overall objective function for long-term time-domain programming is constructed as follows:

[0165] ;

[0166] In the formula, Sampling time The overall objective function value of long-term programming. , , Weighting coefficients for different optimization objectives are used to achieve a balance between minimum energy consumption, trajectory accuracy, and attitude stability.

[0167] 23) Solve the long-term vehicle dynamics programming problem;

[0168] The reference sequence of overall lateral forces and yaw moments of the vehicle in the long time domain is solved using dynamic programming, as follows:

[0169] Based on the Bellman optimality principle, the sampling time is constructed. The optimal cost function of the overall objective function of time-domain programming Its expression is as follows:

[0170] ;

[0171] In the formula, Sampling time The optimal cost function value, It is a feasible set of the overall lateral forces and yaw moments of the vehicle;

[0172] according to The optimal cost function at time step, from stage The process proceeds recursively forward until the optimal cost function for all stages is obtained; this process involves enumerating the feasible set of overall vehicle lateral forces and yaw moments. And by combining the long-term vehicle dynamics planning model, the system state of the next stage can be obtained, thus achieving global search;

[0173] Constructing the optimal policy function Its expression is as follows:

[0174] ;

[0175] In the formula, Sampling time The optimal policy function value;

[0176] Calculations based on long-time domain vehicle dynamics planning model State vector at time step ,as follows:

[0177] ;

[0178] Finally, a reference sequence of the overall lateral force and yaw moment of the vehicle in the long-term domain is obtained, expressed as follows:

[0179] ;

[0180] in, For the first The overall lateral force of the reference vehicle at the predicted step time. For the first The reference vehicle yaw moment at the predicted step.

[0181] 3) Based on the reference sequence of the overall lateral force and yaw moment of the vehicle in the long time domain and the real-time state information of the vehicle, a short-time domain distributed steer-by-wire system angle planning model is constructed, and the four-wheel steering angle reference target that satisfies the system constraints is obtained by solving the model.

[0182] Specifically, step 3) is as follows:

[0183] 31) Construct a short-time-domain distributed steer-by-wire system angle planning model;

[0184] Define the state vector of the steering angle planning model for a short-time distributed steering-by-wire system. ,as follows:

[0185] ;

[0186] Input vector The four wheel corners are selected as follows:

[0187] ;

[0188] In the formula, Sampling time The turning angles of the left front wheel, right front wheel, left rear wheel, and right rear wheel;

[0189] The following is a model for the steering angle planning of a short-time distributed steer-by-wire system:

[0190] ;

[0191] In the formula, for The state vector of the short-time-domain distributed steering-by-wire system's angle planning model. The nonlinear function of the steering angle planning model for a short-time domain distributed steer-by-wire system is detailed below:

[0192] ;

[0193] In the formula, , , , , Sampling time The lateral forces of the left front tire, the right front tire, the left rear tire, and the right rear tire;

[0194] The lateral forces of the tire are described using a linear tire model, as follows:

[0195] ;

[0196] In the formula, , , , These are the lateral stiffness of the left front tire, the right front tire, the left rear tire, and the right rear tire, respectively. , , , Sampling time The slip angles of the left front tire, right front tire, left rear tire, and right rear tire are calculated using the following formulas:

[0197] ;

[0198] 32) Construct the overall objective function for short-time domain programming;

[0199] Constructing a short-time-domain trajectory tracking objective function as follows:

[0200] ;

[0201] In the formula, For the prediction time domain of the steering angle planning model of a short-time distributed steer-by-wire system, For the first Predict the timing, and , For the first Predict the lateral position of the vehicle at any given moment. For the first The reference lateral position of the vehicle at the predicted moment;

[0202] The overall lateral force and yaw moment of the vehicle can be calculated from the lateral forces of the four wheels of the vehicle, as follows:

[0203] ;

[0204] The tracking objective function of the overall lateral force and yaw moment reference sequence of the vehicle for:

[0205] ;

[0206] In the formula, For the first The overall lateral force of the vehicle at the predicted moment. For the first The overall lateral force of the reference vehicle at the predicted step time. For the first Predict the vehicle yaw moment at the next step. For the first Reference vehicle yaw moment at the predicted step;

[0207] Using the corner consistency objective function To improve the synchronization of the distributed steering-by-wire system, the objective function for steering angle consistency is... The expression is as follows:

[0208] ;

[0209] In the formula, For the first The left front wheel steering angle at the predicted moment. For the first The right front wheel steering angle at the predicted moment. For the first The left rear wheel steering angle at the predicted moment. For the first The right rear wheel steering angle at the predicted moment;

[0210] In summary, the overall objective function of short-time domain programming is... It can be written as:

[0211] ;

[0212] In the formula, , , The weighting coefficients are used to correspond to different optimization objectives, in order to achieve a comprehensive balance between trajectory tracking accuracy, cloud reference tracking performance, and corner consistency;

[0213] 33) Construct the steering angle planning constraints for the short-time domain distributed steering-by-wire system;

[0214] Based on the overall objective function of short-time domain programming, input constraints and state constraints are established for the angle planning model of short-time domain distributed steer-by-wire system.

[0215] Set the maximum allowable range of steering angle for the distributed steering-by-wire system as follows:

[0216] ;

[0217] In the formula, For the first Predict the turning angle of each wheel at any given moment. and These are the minimum and maximum allowable turning angles for each wheel, respectively.

[0218] Constraints are imposed on the rate of change of the steering angle to prevent excessively rapid changes in the angle from causing control jitter, as follows:

[0219] ;

[0220] In the formula, This represents the maximum angle change within a single control cycle.

[0221] Constraints are imposed on the sideslip angle and yaw rate to ensure the vehicle maintains good handling stability, as follows:

[0222] ;

[0223] In the formula, For the first The vehicle's sideslip angle at the predicted moment. For the first The vehicle yaw rate at the predicted moment. and These are the maximum permissible values ​​for the sideslip angle and the maximum permissible value for the yaw rate, respectively.

[0224] 34) Solve the angle planning problem for a short-time distributed steering-by-wire system;

[0225] The short-time-domain distributed steer-by-wire system cornering planning model in step 31), the short-time-domain overall objective function in step 32), and the short-time-domain distributed steer-by-wire system cornering planning constraints in step 33) are combined into a finite-time-domain programming problem, which takes the following form:

[0226] ;

[0227] In the formula, Sampling time The predicted four-wheel turn sequence in the time domain is as follows:

[0228] ;

[0229] In the formula, For the first The input vector of the short-time-domain distributed steering-by-wire system angle planning model at the step prediction time;

[0230] The above finite-time programming problem can be represented as a constrained quadratic programming problem. The optimal control sequence can be obtained online using a fast quadratic programming solver or a sequential quadratic programming algorithm, as follows:

[0231] ;

[0232] in, The optimal control sequence obtained by solving the finite-time programming problem. For the first The optimal input vector corresponding to the prediction time step;

[0233] Through the above solution process, the four-wheel steering angle reference target sequence that satisfies the system constraints at the current sampling time is obtained. .

[0234] 4) Based on the four-wheel steering angle reference target that satisfies the system constraints, drive the four steering actuators of the vehicle to generate the corresponding actual steering angles in order to achieve vehicle steering control;

[0235] Specifically, step 4) includes: collecting the actual turning angle of each wheel, constructing a turning angle closed-loop controller, inputting the error between the actual turning angle and the reference turning angle into the PID controller for adjustment, the PID controller calculates the drive control quantity of the steering motor according to the preset proportional term, integral term and derivative term, and drives each steering motor to output the corresponding current, so that the actual turning angle of the wheel gradually approaches the reference turning angle, realizing the tracking control of the four-wheel turning angle, ensuring that the vehicle can accurately track the reference trajectory and achieve optimal global performance.

[0236] This invention has many specific applications. The above description is only a preferred embodiment of this invention. It should be noted that for those skilled in the art, several improvements can be made without departing from the principle of this invention, and these improvements should also be considered within the scope of protection of this invention.

Claims

1. A distributed drive-by-wire steering angle planning system, characterized in that... include: Cloud-based information acquisition unit, cloud-based long-term domain planning unit, vehicle-cloud information interaction unit, vehicle-side information acquisition unit, vehicle-side short-term domain planning unit, and steering actuator control unit; The cloud information acquisition unit is used to acquire vehicle global reference information and send the vehicle global reference information to the cloud long-term domain planning unit and the vehicle-cloud information interaction unit. The vehicle global reference information includes: future driving reference trajectory, future reference vehicle speed, road curvature, and road surface adhesion coefficient; The cloud-based long-term planning unit is used to receive global vehicle reference information obtained by the cloud-based information acquisition unit and real-time vehicle status information sent by the vehicle-cloud information interaction unit, so as to construct a long-term vehicle dynamics planning model, solve for the reference sequence of the overall lateral force and yaw moment of the vehicle in the long-term domain, and send the reference sequence to the vehicle-cloud information interaction unit; the real-time vehicle status information includes: longitudinal vehicle speed, yaw rate, center of gravity sideslip angle, heading angle, and vehicle position. The vehicle-cloud information interaction unit is used to receive real-time vehicle status information collected by the vehicle-side information acquisition unit and send it to the cloud-based long-term domain planning unit, receive the overall vehicle lateral force and yaw moment reference sequence sent by the cloud-based long-term domain planning unit and the vehicle global reference information sent by the cloud-based information acquisition unit, and send them together to the vehicle-side short-term domain planning unit. The vehicle-side information acquisition unit is used to collect real-time vehicle status information and send the real-time vehicle status information to the vehicle-cloud information interaction unit and the vehicle-side short-time domain planning unit. The vehicle-side short-time domain planning unit is used to receive the overall lateral force and yaw moment reference sequence of the vehicle sent by the vehicle-cloud information interaction unit and the real-time vehicle status information sent by the vehicle-side information acquisition unit, construct a short-time domain distributed steer-by-wire system angle planning model, solve for the four-wheel angle reference target that meets the constraints, and send it to the steering actuator control unit. The steering actuator control unit is used to control the vehicle's steering actuator to generate the corresponding actual steering angle based on the four-wheel steering angle reference target sent by the vehicle-side short-time domain planning unit, so as to achieve vehicle steering control.

2. A distributed steering angle planning method based on the system described in claim 1, characterized in that, The method includes the following steps: Step 1) Based on the vehicle's current location, obtain global reference information for the vehicle from the map and road parameter database; Step 2) Using the vehicle's global reference information and real-time vehicle state information as reference targets and initial states, construct a long-term vehicle dynamics planning model. With the optimization objectives of minimizing energy consumption, enhancing vehicle flexibility, and improving lateral stability, solve for the reference sequence of the overall lateral force and yaw moment of the vehicle in the long-term domain. Step 3) Based on the reference sequence of the overall lateral force and yaw moment of the vehicle in the long time domain and the real-time state information of the vehicle, construct the angle planning model of the short time domain distributed steer-by-wire system, and solve for the four-wheel angle reference target that satisfies the system constraints. Step 4) Drive the four steering actuators of the vehicle to generate the corresponding actual steering angles based on the four-wheel steering angle reference target that satisfies the system constraints, so as to achieve vehicle steering control.

3. The distributed drive-by-wire steering angle planning method according to claim 2, characterized in that, Step 2) is as follows: 21) Construct a long-term vehicle dynamics planning model; Define the state vector of the long-time-domain vehicle dynamics planning model; The input vectors of the long-time-domain vehicle dynamics planning model are the overall lateral force and yaw moment of the vehicle; Based on the lateral and yaw motion characteristics of vehicles, a long-term time-domain vehicle dynamics planning model is constructed. 22) Construct the overall objective function for long-term programming; Establish the vehicle's equivalent steering resistance function; Based on the vehicle's equivalent steering resistance function, a long-time domain energy consumption objective function is constructed. Construct a long-term trajectory tracking objective function; Using the vehicle's center of gravity sideslip angle as a key stability indicator, a vehicle stability objective function is constructed. By combining the three objective functions of energy consumption, trajectory tracking, and stability, an overall objective function for long-term time-domain planning is constructed. 23) Solve the long-term vehicle dynamics programming problem; The reference sequence of overall lateral forces and yaw moments of the vehicle in the long time domain is solved using dynamic programming, as follows: Based on the Bellman optimality principle, the sampling time is constructed. The optimal cost function for the overall objective function of time-domain programming; according to The optimal cost function at time step, from stage The process is repeated step by step forward until the optimal cost function for all stages is obtained. This is the prediction time domain for a long-term vehicle dynamics planning model; the process involves enumerating feasible sets of overall vehicle lateral forces and yaw moments. And by combining the long-term vehicle dynamics planning model, the system state of the next stage can be obtained, thus achieving global search; Constructing the optimal policy function ; Calculations based on long-time domain vehicle dynamics planning model State vector at time step ; Finally, a reference sequence of the overall lateral force and yaw moment of the vehicle in the long time domain is obtained.

4. The distributed drive-by-wire steering angle planning method according to claim 3, characterized in that, Step 3) is as follows: 31) Construct a short-time-domain distributed steer-by-wire system angle planning model; Define the state vector of the steering angle planning model for a short-time distributed steering-by-wire system. ; Input vector Select four wheel corners; Construct a short-time-domain distributed steering-by-wire system angle planning model; A linear tire model is used to describe the tire's lateral forces; 32) Construct the overall objective function for short-time domain programming; Construct a short-time-domain trajectory tracking objective function; The overall lateral force and yaw moment of the vehicle are obtained by calculating the lateral forces of the four wheels of the vehicle; The tracking objective function is obtained by obtaining the reference sequence of the overall lateral force and yaw moment of the vehicle; Using the corner consistency objective function To improve the synchronization of the distributed steer-by-wire system; 33) Construct the steering angle planning constraints for the short-time domain distributed steering-by-wire system; Based on the overall objective function of short-time domain programming, input constraints and state constraints are established for the angle planning model of short-time domain distributed steer-by-wire system. Set the maximum allowable range of steering angle for the distributed steering-by-wire system as follows: ； In the formula, For the first Predict the turning angle of each wheel at any given moment. and These are the minimum and maximum allowable turning angles for each wheel, respectively. Constraints are imposed on the rate of change of the steering angle to prevent excessively rapid changes in the steering angle from causing control jitter; Constraints are imposed on the sideslip angle and yaw rate of the center of gravity to ensure that the vehicle maintains good handling stability. 34) Solve the angle planning problem for a short-time distributed steering-by-wire system; The short-time-domain distributed steer-by-wire system cornering planning model in step 31), the short-time-domain overall objective function in step 32), and the short-time-domain distributed steer-by-wire system cornering planning constraints in step 33) are combined into a finite-time-domain programming problem. The finite-time programming problem is represented as a constrained quadratic programming problem, and the optimal control sequence is obtained online using a fast quadratic programming solver or a sequential quadratic programming algorithm. Through the above solution process, the four-wheel steering angle reference target sequence that satisfies the system constraints at the current sampling time is obtained. .

5. The distributed steer-by-wire angle planning method according to claim 4, characterized in that, Step 4) specifically includes: collecting the actual turning angle of each wheel, constructing a turning angle closed-loop controller, inputting the error between the actual turning angle and the reference turning angle into the PID controller for adjustment, the PID controller calculates the drive control quantity of the steering motor according to the preset proportional term, integral term and derivative term, and drives each steering motor to output the corresponding current, so that the actual turning angle of the wheel gradually approaches the reference turning angle, realizing the tracking control of the four-wheel turning angle, ensuring that the vehicle can accurately track the reference trajectory and achieve optimal global performance.

Images