Information physical system oriented to electro-hydraulic drive-by-wire chassis and game optimization method thereof

Abstract

The invention discloses an electro-hydraulic drive-by-wire chassis oriented information physical system and a game optimization method thereof. The system comprises an electro-hydraulic drive-by-wire chassis big data unit, a vehicle and road cloud traffic information unit, an electro-hydraulic drive-by-wire chassis domain control unit, a vehicle condition unit, a road condition unit and an execution unit. The electro-hydraulic drive-by-wire chassis big data unit is in data communication with the vehicle and road cloud traffic information unit and the electro-hydraulic drive-by-wire chassis domain control unit. The vehicle and road cloud traffic information unit is also in data communication with a vehicle condition unit and a road condition unit; the electro-hydraulic drive-by-wire chassis domain control unit is also in data communication with the execution unit; the electro-hydraulic drive-by-wire chassis domain control unit is optimally controlled in an information physical system environment, so that the energy efficiency and the stability of the vehicle are improved, and the driving comfort is considered. The coupling relation between subsystems of the chassis can be processed, and the optimal performance can be achieved through steering coordination and driving control under different working conditions.

Description

Technical Field

[0001] The present invention belongs to the field of control optimization of a drive-by-wire chassis, and in particular relates to an information-physical system for an electro-hydraulic drive-by-wire chassis and a game optimization method thereof. Background Art

[0002] With the development of intelligent network technology and new energy vehicles, the electro-hydraulic wire-controlled chassis system has begun to be used in autonomous vehicles. Due to the limitations of its mechanical structure, the traditional chassis system has certain limitations in control accuracy and response speed. The electro-hydraulic wire-controlled chassis eliminates mechanical connections and uses electronic components such as sensors, electronic control units and motors to achieve higher control freedom and response speed, thereby improving the vehicle's handling performance and safety.

[0003] The optimization of the drive-by-wire chassis system mainly increases the controllability and comfort of the chassis by integrating the chassis system and adding a game system. The Chinese invention patent application number is CN202110225619.2, and the patent name is "A drive-by-wire chassis information-physical system and control method in an intelligent traffic environment". It proposes to integrate the wheel hub motor with the suspension, cancel the traditional structures such as the engine and clutch, simplify the structure of the chassis, and improve active safety and operational stability. However, the patent ignores the personalized needs of users, and its control method is less effective for different driving styles. The Chinese invention patent application number is CN202110471884.9, and the patent name is "Adaptive obstacle avoidance system for intelligent connected vehicles based on drive-by-wire steering and game results". An adaptive obstacle avoidance system for intelligent connected vehicles based on drive-by-wire steering and game results is proposed, and the collected human-vehicle environment information during the game is reported to the central server. After training, the obstacle avoidance ability is improved, but the dynamic change process of the weight factor of the evaluation network is lacking, and its generalization ability and fault tolerance are poor.

[0004] Compared with vehicles equipped with traditional chassis, the electro-hydraulic drive-by-wire chassis system has certain advantages in vehicle handling performance and safety. However, when different subsystems of the electro-hydraulic drive-by-wire chassis complete the same control target, conflicts may occur and affect the overall performance of the chassis. Therefore, it is of great significance to integrate different levels of traffic information physical systems and electro-hydraulic drive-by-wire chassis in actual scenarios, consider the influence of vehicle stability constraints, energy, etc. to optimize the electro-hydraulic drive-by-wire chassis game, and give full play to the control performance of the electro-hydraulic drive-by-wire chassis. Summary of the invention

[0005] In view of the above-mentioned deficiencies of the prior art, the purpose of the present invention is to provide a cyber-physical system for an electro-hydraulic wire-controlled chassis and a game optimization method thereof, which can optimize the control of the electro-hydraulic wire-controlled chassis domain control unit in a cyber-physical system environment to improve the energy efficiency and stability of the vehicle while taking into account driving comfort. The method of the present invention can not only handle the coupling relationship between the various subsystems of the chassis, but also achieve optimal performance by coordinating steering and drive control under different working conditions.

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

[0007] A cyber-physical system for an electro-hydraulic wire-controlled chassis of the present invention comprises: an electro-hydraulic wire-controlled chassis big data unit, a vehicle-road cloud traffic information unit, an electro-hydraulic wire-controlled chassis domain control unit, a vehicle condition unit, a road condition unit and an execution unit; the electro-hydraulic wire-controlled chassis big data unit communicates data with the vehicle-road cloud traffic information unit and the electro-hydraulic wire-controlled chassis domain control unit respectively; the vehicle-road cloud traffic information unit also communicates data with the vehicle condition unit and the road condition unit; the electro-hydraulic wire-controlled chassis domain control unit also communicates data with the execution unit;

[0008] The vehicle condition unit comprises: a vehicle state storage module and a vehicle condition sensor group module;

[0009] The vehicle condition sensor group module is used to collect vehicle condition data and send the collected data to the vehicle condition storage module;

[0010] A vehicle status storage module, used to receive and store data sent by the vehicle condition sensor group module, and send the data to the first information storage module in the vehicle-road-cloud traffic information unit;

[0011] The road condition unit comprises: a road state storage module and a road condition sensor group module;

[0012] A road condition sensor group module is used to collect road condition data and send the collected data to the road condition storage module;

[0013] A road state storage module, used to receive and store data sent by the road condition sensor group module, and send the data to the first information storage module in the vehicle-road-cloud traffic information unit;

[0014] The vehicle-road-cloud traffic information unit comprises: a first information storage module and a game optimization module;

[0015] The first information storage module performs data transmission and data storage with the electric-hydraulic wire-controlled chassis big data unit, the vehicle condition unit, the road condition unit and the game optimization module respectively;

[0016] A game optimization module, used for performing game optimization on the active front wheel steering and the direct yaw moment according to the data in the first information storage module, wherein the optimization result is the actual output of the active front wheel steering and the actual output of the direct yaw moment, and the optimization result is sent to the electro-hydraulic wire-controlled chassis big data unit;

[0017] The electro-hydraulic wire-controlled chassis big data unit comprises: a data management module, a machine learning module and an evaluation network module;

[0018] A data management module, which transmits and stores data with the first information storage module, the electro-hydraulic wire-controlled chassis domain control unit, the evaluation network module and the machine learning module respectively;

[0019] An evaluation network module is used to estimate the coordination state of the evaluation network of the data collected by the vehicle condition sensor group module and the road condition sensor group module, calculate the target coordination state of the data, make a difference between the estimated result of the coordination state of the evaluation network and the target coordination state, obtain the error function of the evaluation network, use the gradient descent algorithm to optimize the error function of the evaluation network, and send the optimization result to the data management module and the first information storage module;

[0020] A machine learning module, used to judge the driving condition through a decision tree model learned from an ideal operation database based on the data collected by the vehicle condition sensor group module and the road condition sensor group module, optimize the model simultaneously, and transmit the vehicle driving condition data to the data management module and the first information storage module;

[0021] The electro-hydraulic wire-controlled chassis domain control unit comprises: a second information storage module and an execution network module;

[0022] The second information storage module performs data transmission and data storage with the data management module, the execution unit, and the execution network module respectively;

[0023] an execution network module, for generating a control instruction using a BP neural network according to the data of the actual output of the active front wheel steering and the actual output of the direct yaw moment in the second information storage module, and sending the control instruction to the second information storage module;

[0024] The execution unit includes: a wire-controlled chassis steering module and a wheel hub motor drive module;

[0025] A wire-controlled chassis steering module, used to perform active front wheel steering according to control instructions generated by the execution network module;

[0026] The wheel hub motor drive module is used to execute direct yaw torque according to the control instructions generated by the execution network module.

[0027] Furthermore, the decision tree model is used for classification and regression tasks; data in an ideal operation database is read; according to the Gini index method, a feature is selected as a node splitting criterion, and the data in the ideal operation database is split into subsets according to the selected feature; the Gini index method is repeated for each subset, the best feature is selected for splitting, and the data is continued to be split until the stopping condition is met; for new vehicle status data and road status data, starting from the root node, according to the comparison result of the feature value, until a leaf node is judged, and the value of the leaf node is the driving condition result.

[0028] Furthermore, the Gini index method measures the impurity of a data set; the smaller the Gini index, the purer the data set, and the feature with the smallest Gini index is selected for splitting.

[0029] Furthermore, the stopping conditions include: reaching a preset maximum depth of the tree, the number of samples contained in each leaf node is less than a preset minimum number of samples, and the data set no longer has significant splits.

[0030] Furthermore, the ideal operation database is an offline synchronous database, which contains a large amount of historical data under different driving conditions. It stores the optimal operation and control parameters of the vehicle under different driving conditions, which are derived from ideal operation data collected during experiments, simulations and actual driving.

[0031] Furthermore, the driving conditions include: urban driving conditions, rural road driving conditions, special road driving conditions and mixed driving conditions.

[0032] Furthermore, the data collected by the vehicle condition sensor group module include: vehicle slip rate, center of mass sideslip angle, longitudinal vehicle speed, yaw angular velocity, longitudinal acceleration speed, and lateral acceleration.

[0033] Furthermore, the data collected by the road condition sensor group module includes: the number and position of other vehicles around the vehicle, the ground adhesion coefficient and the road slope.

[0034] Furthermore, the control-by-wire chassis steering module includes: a steering motor controller, a steering actuator motor, a gear rack and a wheel; the steering motor controller controls the steering actuator motor through control instructions, and the steering actuator motor is connected to the wheel through a gear rack; the steering motor controller receives the control instructions sent by the execution network module, drives the steering actuator motor to generate torque, and the torque is transmitted to the wheel under the action of the gear rack to realize the vehicle steering operation and complete the active front wheel steering.

[0035] Furthermore, the hub motor drive module includes: a drive motor controller and a hub motor; the drive motor controller controls the hub motor through control instructions, and the hub motor is mechanically connected to the wheel; the drive motor controller receives the control instructions generated by the execution network module, drives the hub motor to generate driving force, and the hub motor generates a yaw torque to complete the direct yaw torque.

[0036] The present invention also provides a game optimization method for a cyber-physical system for an electro-hydraulic wire-controlled chassis. Based on the above system, the steps are as follows:

[0037] 1) Collect vehicle status data and road status data;

[0038] 2) Estimate the co-state of the evaluation network of the data collected in step 1), calculate the target co-state of the data, make a difference between the estimated result of the co-state of the evaluation network and the target co-state, obtain the error function of the evaluation network, and optimize the error function to obtain the optimization result;

[0039] 3) The data collected in step 1) is used to determine the driving condition to obtain vehicle driving condition data;

[0040] 4) using the data collected in step 1), the optimization result in step 2), and the vehicle driving condition data in step 3) as game optimization to obtain the actual output of the active front wheel steering and the actual output of the direct yaw moment;

[0041] 5) using the actual output of the active front wheel steering and the actual output of the direct yaw moment obtained in step 4), the control instructions of the active front wheel steering and the direct yaw moment are calculated;

[0042] 6) Complete vehicle control according to the control instruction obtained in step 5).

[0043] Furthermore, the estimation of the coordination state of the evaluation network in step 2) is expressed as:

[0044]

[0045] In the formula, is X(k),ω AFS The estimation of the co-state equation for the active front wheel steering at the l+1th iteration in the state, is X(k),ω DYC The estimate of the co-state equation for the l+1th iteration of the direct yaw moment in the state, is the state quantity after augmentation at time k, ω AFS is the weight coefficient of the middle layer of the active front wheel steering neural network, ω DYC is the weight coefficient of the intermediate layer of the direct yaw moment neural network, ω AFS(l+1)is the weight coefficient of the direct yaw moment for the l+1th iteration, ω DYC(l+1) is the weight coefficient of the direct yaw moment for the l+1th iteration, φ AFS(l+1) is the basis function of the active front wheel steering at the l+1th iteration, φ DYC(l+1) is the basis function of the direct yaw moment for the l+1th iteration.

[0046] Furthermore, the expression of the target cooperative state in step 2) is as follows:

[0047]

[0048] Where D AFS is the weight matrix of active front wheel steering, D DYC is the weight matrix of the direct yaw moment, Q AFS is the tracking error of the active front wheel steering, Q DYC is the tracking error of the direct yaw moment, λ AFS(l+1) (X(k)) is the co-state equation of the l+1th iteration of active front wheel steering under the X(k) state, λ DYC(l+1) (X(k)) is the co-state equation of the l+1th iteration of the direct yaw moment under the X(k) state, γ AFS is the discount factor of the active front steering performance index, γ DYC is the discount factor for the direct yaw moment performance indicator.

[0049] Furthermore, the expression of the error function of the evaluation network in step 2) is:

[0050]

[0051] In the formula, e λAFS(l+1) (k) is the error function of the l+1th iteration of active front wheel steering at time k, e λDYC(l+1) (k) is the error function of the l+1th iteration of the direct yaw moment at time k.

[0052] Furthermore, the specific steps of step 4) are as follows:

[0053] 41) using the data collected in step 1), the optimization results in step 2), and the vehicle driving condition data in step 3), calculate the state output equation of the closed-loop Nash equilibrium and the state equation of the reference signal system;

[0054] 42) Calculate the tracking error e of the active front wheel steering AFS (k) and the tracking error of the direct yaw moment e DYC (k);

[0055] 43) Calculating a cost function to enable the vehicle to track the reference trajectory with minimal energy consumption;

[0056] 44) Calculate the state space equation of the augmented game system and the performance index function of the augmented game system;

[0057] 45) Calculate the game output of the weight factor.

[0058] Furthermore, the state output equation of the closed-loop Nash equilibrium in step 41) is:

[0059]

[0060] In the formula, x(k+1) is the state quantity at time k+1, x(k) is the state quantity at time k, and u AFS (k) is the input of active front wheel steering at time k, u DYC (k) is the input of the direct yaw moment at time k, A is the coefficient matrix of the state quantity, B AFS is the coefficient matrix of the active front wheel steering input, B DYC is the coefficient matrix of the direct yaw moment input, C AFS is the output matrix of active front wheel steering, C DYC is the output matrix of the direct yaw moment, y AFS (k) is the actual output of active front wheel steering at time k, y DYC (k) is the actual output of the direct yaw moment at time k;

[0061] The state equation of the reference signal system is:

[0062]

[0063] In the formula, is the state of the reference signal system at time k+1, is the state of the reference signal system at time k, r AFS (k) is the expected output of the active front wheel steering reference signal system at time k, r DYC (k) is the expected output of the direct yaw moment reference signal system at time k, F is the system matrix corresponding to the reference signal system, G AFS is the output matrix of the active front wheel steering reference signal system matrix, G DYC is the output matrix of the direct yaw moment reference signal system matrix.

[0064] Furthermore, in step 42), the tracking error of the active front wheel steering and the direct yaw moment is defined as follows:

[0065]

[0066] In the formula, e AFS(k) is the tracking error of the active front wheel steering at time k, e DYC (k) is the tracking error of the direct yaw moment at time k.

[0067] Furthermore, the cost function in step 43) is:

[0068]

[0069] in, for The cost function of active front wheel steering in the state, for The cost function of the direct yaw moment in the state, r AFS (e AFS (i),u AFS (i)) is e AFS (i),u AFS (i) The expected output of the active front wheel steering reference signal system, r DYC (e DYC (i),u DYC (i)) is e DYC (i),u DYC (i) The expected output of the direct yaw moment reference signal system at time k, r AFS (e AFS (i),u AFS (i)) and r DYC (e DYC (i),u DYC (i)) is expressed as follows:

[0070]

[0071] In the formula, R AFS is the control cost weight matrix of active front wheel steering, R DYC is the control cost weight matrix of the direct yaw moment.

[0072] Furthermore, the state space equation of the augmented game system and the performance index function of the augmented game system in step 44) are as follows:

[0073]

[0074]

[0075] In the formula, is the state quantity after augmentation at time k+1, is the state quantity after augmentation at time k, is the coefficient matrix of the augmented state quantity, is the coefficient matrix of the augmented active front wheel steering input, is the coefficient matrix of the augmented direct yaw moment input, J AFS (X(k)) is the cost function of active front wheel steering under the X(k) state, J DYC (X(k)) is the cost function of the direct yaw moment in the X(k) state.

[0076] Furthermore, in step 45), according to the dynamic non-cooperative game theory, the closed-loop Nash strategy This holds if and only if the following recurrence relation holds:

[0077]

[0078] in:

[0079]

[0080] In the formula, J is the time k AFS (X(k)) intermediate variables, J is the time k DYC (X(k)) intermediate variables, is the optimal value of active front wheel steering at time k, is the optimal value of the direct yaw moment at time k.

[0081] Furthermore, the step 5) specifically includes:

[0082] 51) calculating an estimated input amount of the active front wheel steering for the first iteration and an estimated input amount of the direct yaw moment for the first iteration according to the actual output of the active front wheel steering and the actual output of the direct yaw moment output in step 4);

[0083] 52) Calculate the target value of the execution network for the lth iteration;

[0084] 53) Subtract the input estimate of the lth iteration from the target value of the execution network to obtain the error function of the execution network;

[0085] 54) Adopting the adaptive rule of gradient descent, updating the weights of the execution network, and generating the best active front wheel steering and direct yaw moment control instructions, and completing the control of the vehicle according to the control instructions.

[0086] Further, in step 51), according to the actual output of the active front wheel steering and the actual output of the direct yaw moment output in step 4), the input amount estimation of the lth iteration of the active front wheel steering and the input amount estimation of the lth iteration of the direct yaw moment are calculated:

[0087]

[0088] in, is X(k),θ AFS The input estimation of the active front wheel steering in the lth iteration under the state, is X(k),θ DYC The input quantity estimation of the direct yaw moment at the lth iteration in the state, θ AFS is the weight coefficient of the middle layer of the active front wheel steering neural network, θ DYC is the weight coefficient of the intermediate layer of the direct yaw moment neural network, ψ AFS(l) (X(k)) is the basis function of the active front wheel steering at the lth iteration under the X(k) state, ψ DYC(l) (X(k)) is the basis function of the lth iteration of the direct yaw moment in the X(k) state.

[0089] Furthermore, the target value of the network's first iteration in step 52) is expressed as follows:

[0090]

[0091] Among them, u AFS(l) (X(k)) is the input of the active front wheel steering at the lth iteration under the X(k) state; u DYC(l) (X(k)) is the input of the direct yaw moment at the lth iteration under the X(k) state; AFS(l) (X(k+1)) is the co-state equation of the first iteration of active front wheel steering under the X(k+1) state, λ DYC(l) (X(k+1)) is the co-state equation of the lth iteration of the direct yaw moment in the X(k+1) state.

[0092] Furthermore, in step 53), according to the target value of the execution network, the error function of the execution network is obtained as follows:

[0093]

[0094] Among them, e uAFS(l) (k) is the target value of the active front wheel steering execution network for the first iteration, e uDYC(l) (k) is the target value of the direct yaw moment execution network at the lth iteration.

[0095] Beneficial effects of the present invention:

[0096] The present invention uses the execution network module in the electro-hydraulic wire-controlled chassis domain control unit to read the actual output of the active front wheel steering and the actual output of the direct yaw torque in the second information storage module, and calculates the optimal control instructions of the wire-controlled chassis steering module and the wheel hub motor drive module through the BP neural network structure. The neural network has a strong fitting ability and can comprehensively consider multiple complex constraints, so that the vehicle has a suitable control scheme.

[0097] The present invention solves the problem of conflicts and prisoners' dilemma when the control unit of the electro-hydraulic wire-controlled chassis completes multiple control targets through the game optimization module in the vehicle-road-cloud traffic information unit, thereby ensuring the comprehensive performance of the chassis.

[0098] The present invention uses the evaluation network module in the electro-hydraulic wire-controlled chassis big data unit and adopts the learning rule of the steepest descent method to dynamically send feedback information to the game optimization module, so that the game optimization module has better generalization and fault tolerance capabilities, solves the dimensionality disaster problem in dynamic programming, and provides a practical theory and method for the optimal control problem of the high-dimensional complex nonlinear system of the electro-hydraulic wire-controlled chassis. BRIEF DESCRIPTION OF THE DRAWINGS

[0099] Figure 1 The figure is a functional block diagram of the system of the present invention.

[0100] Figure 2 Flow chart of the method of the present invention. DETAILED DESCRIPTION

[0101] In order to facilitate the understanding of those skilled in the art, the present invention is further described below in conjunction with embodiments and drawings. The contents mentioned in the implementation modes are not intended to limit the present invention.

[0102] Reference Figure 1 As shown, a cyber-physical system for an electro-hydraulic wire-controlled chassis of the present invention comprises: an electro-hydraulic wire-controlled chassis big data unit, a vehicle-road cloud traffic information unit, an electro-hydraulic wire-controlled chassis domain control unit, a vehicle condition unit, a road condition unit and an execution unit; the electro-hydraulic wire-controlled chassis big data unit communicates data with the vehicle-road cloud traffic information unit and the electro-hydraulic wire-controlled chassis domain control unit respectively; the vehicle-road cloud traffic information unit also communicates data with the vehicle condition unit and the road condition unit; the electro-hydraulic wire-controlled chassis domain control unit also communicates data with the execution unit;

[0103] The vehicle condition unit comprises: a vehicle state storage module and a vehicle condition sensor group module;

[0104] The vehicle condition sensor group module is used to collect vehicle condition data and send the collected data to the vehicle condition storage module;

[0105] A vehicle status storage module, used to receive and store data sent by the vehicle condition sensor group module, and send the data to the first information storage module in the vehicle-road-cloud traffic information unit;

[0106] The road condition unit comprises: a road state storage module and a road condition sensor group module;

[0107] A road condition sensor group module is used to collect road condition data and send the collected data to the road condition storage module;

[0108] A road state storage module, used to receive and store data sent by the road condition sensor group module, and send the data to the first information storage module in the vehicle-road-cloud traffic information unit;

[0109] The vehicle-road-cloud traffic information unit comprises: a first information storage module and a game optimization module;

[0110] The first information storage module performs data transmission and data storage with the electric-hydraulic wire-controlled chassis big data unit, the vehicle condition unit, the road condition unit and the game optimization module respectively;

[0111] A game optimization module, used for performing game optimization on the active front wheel steering and the direct yaw moment according to the data in the first information storage module, wherein the optimization result is the actual output of the active front wheel steering and the actual output of the direct yaw moment, and the optimization result is sent to the electro-hydraulic wire-controlled chassis big data unit;

[0112] The electro-hydraulic wire-controlled chassis big data unit comprises: a data management module, a machine learning module and an evaluation network module;

[0113] A data management module, which transmits and stores data with the first information storage module, the electro-hydraulic wire-controlled chassis domain control unit, the evaluation network module and the machine learning module respectively;

[0114] An evaluation network module is used to estimate the coordination state of the evaluation network of the data collected by the vehicle condition sensor group module and the road condition sensor group module, calculate the target coordination state of the data, make a difference between the estimated result of the coordination state of the evaluation network and the target coordination state, obtain the error function of the evaluation network, use the gradient descent algorithm to optimize the error function of the evaluation network, and send the optimization result to the data management module and the first information storage module;

[0115] A machine learning module, used to judge the driving condition through a decision tree model learned from an ideal operation database based on the data collected by the vehicle condition sensor group module and the road condition sensor group module, optimize the model simultaneously, and transmit the vehicle driving condition data to the data management module and the first information storage module;

[0116] The decision tree model is used for classification and regression tasks; data in an ideal operation database is read; according to the Gini index method, a feature is selected as a node splitting criterion, and the data in the ideal operation database is split into subsets according to the selected feature; the Gini index method is repeated for each subset, the best feature is selected for splitting, and the data is continued to be split until the stopping condition is met; for new vehicle status data and road status data, starting from the root node, according to the comparison result of the feature value, until a leaf node is judged, and the value of the leaf node is the driving condition result.

[0117] The Gini index method measures the impurity of a data set; the smaller the Gini index, the purer the data set, and the feature with the smallest Gini index is selected for splitting.

[0118] The stopping conditions include: reaching a preset maximum depth of the tree, the number of samples contained in each leaf node is less than a preset minimum number of samples, and the data set no longer has significant splits.

[0119] The ideal operation database is an offline synchronous database, which contains a large amount of historical data under different driving conditions. It stores the optimal operation and control parameters of the vehicle under different driving conditions, which are derived from ideal operation data collected during experiments, simulations and actual driving.

[0120] The electro-hydraulic wire-controlled chassis domain control unit comprises: a second information storage module and an execution network module;

[0121] The second information storage module performs data transmission and data storage with the data management module, the execution unit, and the execution network module respectively;

[0122] an execution network module, for generating a control instruction using a BP neural network according to the data of the actual output of the active front wheel steering and the actual output of the direct yaw moment in the second information storage module, and sending the control instruction to the second information storage module;

[0123] The execution unit includes: a wire-controlled chassis steering module and a wheel hub motor drive module;

[0124] A wire-controlled chassis steering module, used to perform active front wheel steering according to control instructions generated by the execution network module;

[0125] The wheel hub motor drive module is used to execute direct yaw torque according to the control instructions generated by the execution network module.

[0126] Specifically, the driving conditions include: urban driving conditions, rural road driving conditions, special road driving conditions and mixed driving conditions.

[0127] Specifically, the data collected by the vehicle condition sensor group module include: vehicle slip rate, center of mass sideslip angle, longitudinal vehicle speed, yaw angular velocity, longitudinal acceleration speed, and lateral acceleration.

[0128] Specifically, the data collected by the road condition sensor group module include: the number and position of other vehicles around the vehicle, the ground adhesion coefficient and the road slope.

[0129] Specifically, the control-by-wire chassis steering module includes: a steering motor controller, a steering actuator motor, a gear rack and a wheel; the steering motor controller controls the steering actuator motor through control instructions, and the steering actuator motor is connected to the wheel through a gear rack; the steering motor controller receives the control instructions sent by the execution network module, drives the steering actuator motor to generate torque, and the torque is transmitted to the wheel under the action of the gear rack to realize the vehicle steering operation and complete the active front wheel steering.

[0130] Specifically, the hub motor drive module includes: a drive motor controller and a hub motor; the drive motor controller controls the hub motor through control instructions, and the hub motor is mechanically connected to the wheel; the drive motor controller receives the control instructions generated by the execution network module, drives the hub motor to generate driving force, and the hub motor generates yaw torque to complete direct yaw torque.

[0131] Reference Figure 2 As shown, the present invention also provides a game optimization method for a cyber-physical system for an electro-hydraulic wire-controlled chassis. Based on the above system, the steps are as follows:

[0132] 1) Collect vehicle status data and road status data;

[0133] Specifically, the road status data includes: urban driving conditions, rural road driving conditions, special road driving conditions and mixed driving conditions.

[0134] Specifically, the vehicle status data includes: vehicle slip rate, center of mass sideslip angle, longitudinal vehicle speed, yaw angular velocity, longitudinal acceleration speed, and lateral acceleration.

[0135] 2) Estimate the co-state of the evaluation network of the data collected in step 1), calculate the target co-state of the data, make a difference between the estimated result of the co-state of the evaluation network and the target co-state, obtain the error function of the evaluation network, and optimize the error function to obtain the optimization result;

[0136] The estimation of the co-state of the evaluation network is expressed as:

[0137]

[0138] In the formula, is X(k),ω AFSThe estimation of the co-state equation for the active front wheel steering at the l+1th iteration in the state, is X(k),ω DYC The estimate of the co-state equation for the l+1th iteration of the direct yaw moment in the state, is the state quantity after augmentation at time k, ω AFS is the weight coefficient of the middle layer of the active front wheel steering neural network, ω DYC is the weight coefficient of the intermediate layer of the direct yaw moment neural network, ω AFS(l+1) is the weight coefficient of the direct yaw moment for the l+1th iteration, ω DYC(l+1) is the weight coefficient of the direct yaw moment for the l+1th iteration, φ AFS(l+1) is the basis function of the active front wheel steering at the l+1th iteration, φ DYC(l+1) is the basis function of the direct yaw moment for the l+1th iteration.

[0139] The expression of the target co-state is as follows:

[0140]

[0141] Where D AFS is the weight matrix of active front wheel steering, D DYC is the weight matrix of the direct yaw moment, Q AFS is the tracking error of the active front wheel steering, Q DYC is the tracking error of the direct yaw moment, λ AFS(l+1) (X(k)) is the co-state equation of the l+1th iteration of active front wheel steering under the X(k) state, λ DYC(l+1) (X(k)) is the co-state equation of the l+1th iteration of the direct yaw moment under the X(k) state, γ AFS is the discount factor of the active front steering performance index, γ DYC is the discount factor for the direct yaw moment performance indicator.

[0142] The expression of the error function of the evaluation network is:

[0143]

[0144] In the formula, e λAFS(l+1) (k) is the error function of the l+1th iteration of active front wheel steering at time k, e λDYC(l+1) (k) is the error function of the l+1th iteration of the direct yaw moment at time k.

[0145] 3) Using the data collected in step 1) to determine the driving condition using a decision tree model to obtain vehicle driving condition data;

[0146] The decision tree model is used for classification and regression tasks; read the data in the ideal operation database; according to the Gini index method, select a feature as the node splitting criterion, and split the data in the ideal operation database into subsets according to the selected feature; repeat the Gini index method for each subset, select the best feature for splitting, and continue to split the data until the stopping condition is met; for new vehicle status data and road status data, start from the root node and compare the feature values ​​until the leaf node is judged, and the value of the leaf node is the driving condition result.

[0147] 4) Using the data collected in step 1), the optimization results in step 2), and the vehicle driving condition data in step 3) as game optimization, the actual output of the active front wheel steering and the actual output of the direct yaw moment are obtained; the specific steps are as follows:

[0148] 41) using the data collected in step 1), the optimization results in step 2), and the vehicle driving condition data in step 3), calculate the state output equation of the closed-loop Nash equilibrium and the state equation of the reference signal system;

[0149] 42) Calculate the tracking error e of the active front wheel steering AFS (k) and the tracking error of the direct yaw moment e DYC (k);

[0150] 43) Calculating a cost function to enable the vehicle to track the reference trajectory with minimal energy consumption;

[0151] 44) Calculate the state space equation of the augmented game system and the performance index function of the augmented game system;

[0152] 45) Calculate the game output of the weight factor.

[0153] The state output equation of the closed-loop Nash equilibrium in step 41) is:

[0154]

[0155] In the formula, x(k+1) is the state quantity at time k+1, x(k) is the state quantity at time k, and u AFS (k) is the input of active front wheel steering at time k, u DYC (k) is the input of the direct yaw moment at time k, A is the coefficient matrix of the state quantity, B AFS is the coefficient matrix of the active front wheel steering input, B DYC is the coefficient matrix of the direct yaw moment input, C AFS is the output matrix of active front wheel steering, C DYC is the output matrix of the direct yaw moment, y AFS(k) is the actual output of active front wheel steering at time k, y DYC (k) is the actual output of the direct yaw moment at time k;

[0156] The state equation of the reference signal system is:

[0157]

[0158] In the formula, is the state of the reference signal system at time k+1, is the state of the reference signal system at time k, r AFS (k) is the expected output of the active front wheel steering reference signal system at time k, r DYC (k) is the expected output of the direct yaw moment reference signal system at time k, F is the system matrix corresponding to the reference signal system, G AFS is the output matrix of the active front wheel steering reference signal system matrix, G DYC is the output matrix of the direct yaw moment reference signal system matrix.

[0159] Wherein, the tracking error of the active front wheel steering and the direct yaw moment is defined in step 42) as follows:

[0160]

[0161] In the formula, e AFS (k) is the tracking error of the active front wheel steering at time k, e DYC (k) is the tracking error of the direct yaw moment at time k.

[0162] Wherein, the cost function in step 43) is:

[0163]

[0164] in, for The cost function of active front wheel steering in the state, for The cost function of the direct yaw moment in the state, r AFS (e AFS (i),u AFS (i)) is e AFS (i),u AFS (i) The expected output of the active front wheel steering reference signal system, r DYC (e DYC (i),u DYC (i)) is e DYC (i),u DYC (i) The expected output of the direct yaw moment reference signal system at time k, rAFS (e AFS (i),u AFS (i)) and r DYC (e DYC (i),u DYC (i)) is expressed as follows:

[0165]

[0166] In the formula, R AFS is the control cost weight matrix of active front wheel steering, R DYC is the control cost weight matrix of the direct yaw moment.

[0167] Wherein, the state space equation of the augmented game system and the performance index function of the augmented game system in step 44) are as follows:

[0168]

[0169]

[0170] In the formula, is the state quantity after augmentation at time k+1, is the state quantity after augmentation at time k, is the coefficient matrix of the augmented state quantity, is the coefficient matrix of the augmented active front wheel steering input, is the coefficient matrix of the augmented direct yaw moment input, J AFS (X(k)) is the cost function of active front wheel steering under the X(k) state, J DYC (X(k)) is the cost function of the direct yaw moment in the X(k) state.

[0171] Wherein, in step 45), according to the dynamic non-cooperative game theory, the closed-loop Nash strategy This holds if and only if the following recurrence relation holds:

[0172]

[0173] in:

[0174]

[0175] In the formula, J is the time k AFS (X(k)) intermediate variables, J is the time k DYC (X(k)) intermediate variables, is the optimal value of active front wheel steering at time k, is the optimal value of the direct yaw moment at time k.

[0176] 5) Calculating the control instructions of the active front wheel steering and the direct yaw moment using the actual output of the active front wheel steering and the actual output of the direct yaw moment obtained in step 4); specifically including:

[0177] 51) calculating an estimated input amount of the active front wheel steering for the first iteration and an estimated input amount of the direct yaw moment for the first iteration according to the actual output of the active front wheel steering and the actual output of the direct yaw moment output in step 4);

[0178] 52) Calculate the target value of the execution network for the lth iteration;

[0179] 53) Subtract the input estimate of the lth iteration from the target value of the execution network to obtain the error function of the execution network;

[0180] 54) Adopting the adaptive rule of gradient descent, updating the weights of the execution network, and generating the best active front wheel steering and direct yaw moment control instructions, and completing the control of the vehicle according to the control instructions.

[0181] Wherein, in step 51), according to the actual output of the active front wheel steering and the actual output of the direct yaw moment output in step 4), the input amount estimation of the lth iteration of the active front wheel steering and the input amount estimation of the lth iteration of the direct yaw moment are calculated:

[0182]

[0183] in, is X(k),θ AFS The input estimation of the active front wheel steering in the lth iteration under the state, is X(k),θ DYC The input quantity estimation of the direct yaw moment at the lth iteration in the state, θ AFS is the weight coefficient of the middle layer of the active front wheel steering neural network, θ DYC is the weight coefficient of the intermediate layer of the direct yaw moment neural network, ψ AFS(l) (X(k)) is the basis function of the active front wheel steering at the lth iteration under the X(k) state, ψ DYC(l) (X(k)) is the basis function of the lth iteration of the direct yaw moment in the X(k) state.

[0184] Wherein, the target value of the network's first iteration in step 52) is expressed by the following formula:

[0185]

[0186] Among them, u AFS(l)(X(k)) is the input of the active front wheel steering at the lth iteration under the X(k) state; u DYC(l) (X(k)) is the input of the direct yaw moment at the lth iteration under the X(k) state; AFS(l) (X(k+1)) is the co-state equation of the first iteration of active front wheel steering under the X(k+1) state, λ DYC(l) (X(k+1)) is the co-state equation of the lth iteration of the direct yaw moment in the X(k+1) state.

[0187] Wherein, in step 53), according to the target value of the execution network, the error function of the execution network is obtained as follows:

[0188]

[0189] Among them, e uAFS ( l) (k) is the target value of the active front wheel steering execution network for the first iteration, e uDYC ( l) (k) is the target value of the direct yaw moment execution network at the lth iteration.

[0190] 6) Complete vehicle control according to the control instruction obtained in step 5).

[0191] The present invention has many specific application paths. The above is only a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements can be made without departing from the principle of the present invention. These improvements should also be regarded as the protection scope of the present invention.

Claims

1. A cyber-physical system for an electro-hydraulic wire-controlled chassis, characterized in that: include: Electro-hydraulic wire-controlled chassis big data unit, vehicle-road-cloud traffic information unit, electro-hydraulic wire-controlled chassis domain control unit, vehicle condition unit, road condition unit and execution unit; The vehicle condition unit comprises: a vehicle state storage module and a vehicle condition sensor group module; The vehicle condition sensor group module is used to collect vehicle condition data and send the collected data to the vehicle condition storage module; A vehicle status storage module, used to receive and store data sent by the vehicle condition sensor group module, and send the data to the first information storage module in the vehicle-road-cloud traffic information unit; The road condition unit comprises: a road state storage module and a road condition sensor group module; A road condition sensor group module is used to collect road condition data and send the collected data to the road condition storage module; A road state storage module, used to receive and store data sent by the road condition sensor group module, and send the data to the first information storage module in the vehicle-road-cloud traffic information unit; The vehicle-road-cloud traffic information unit comprises: a first information storage module and a game optimization module; The first information storage module performs data transmission and data storage with the electric-hydraulic wire-controlled chassis big data unit, the vehicle condition unit, the road condition unit and the game optimization module respectively; The game optimization module is used to perform game optimization on the active front wheel steering and the direct yaw moment. The optimization results are the actual output of the active front wheel steering and the actual output of the direct yaw moment, and are sent to the electro-hydraulic wire-controlled chassis big data unit; The electro-hydraulic wire-controlled chassis big data unit comprises: a data management module, a machine learning module and an evaluation network module; A data management module, which transmits and stores data with the first information storage module, the electro-hydraulic wire-controlled chassis domain control unit, the evaluation network module and the machine learning module respectively; An evaluation network module is used to estimate the coordination state of the evaluation network of the data collected by the vehicle condition sensor group module and the road condition sensor group module, calculate the target coordination state of the data, and make a difference between the estimated result of the coordination state of the evaluation network and the target coordination state to obtain an error function of the evaluation network, optimize the error function of the evaluation network, and send the optimization result to the data management module and the first information storage module; A machine learning module, used to determine the driving condition and transmit the vehicle driving condition data to the data management module and the first information storage module; The electro-hydraulic wire-controlled chassis domain control unit comprises: a second information storage module and an execution network module; The second information storage module performs data transmission and data storage with the data management module, the execution unit, and the execution network module respectively; an execution network module, for generating a control instruction according to the data of the actual output of the active front wheel steering and the actual output of the direct yaw moment in the second information storage module, and sending the control instruction to the second information storage module; The execution unit includes: a wire-controlled chassis steering module and a wheel hub motor drive module; A wire-controlled chassis steering module, used to perform active front wheel steering according to control instructions generated by the execution network module; The wheel hub motor drive module is used to execute direct yaw torque according to the control instructions generated by the execution network module.

2. A game optimization method for a cyber-physical system for an electro-hydraulic wire-controlled chassis, based on the system of claim 1, characterized in that: The steps are as follows: 1) Collect vehicle status data and road status data; 2) Estimate the co-state of the evaluation network of the data collected in step 1), calculate the target co-state of the data, make a difference between the estimated result of the co-state of the evaluation network and the target co-state, obtain the error function of the evaluation network, and optimize the error function to obtain the optimization result; 3) The data collected in step 1) is used to determine the driving condition to obtain vehicle driving condition data; 4) using the data collected in step 1), the optimization result in step 2), and the vehicle driving condition data in step 3) as game optimization to obtain the actual output of the active front wheel steering and the actual output of the direct yaw moment; 5) using the actual output of the active front wheel steering and the actual output of the direct yaw moment obtained in step 4), the control instructions of the active front wheel steering and the direct yaw moment are calculated; 6) Complete vehicle control according to the control instruction obtained in step 5).

3. The method according to claim 2, characterized in that The estimation of the cooperative state of the evaluation network in step 2) is expressed as: In the formula, is X(k),ω AFS The estimation of the co-state equation for the active front wheel steering at the l+1th iteration in the state, is X(k),ω DYC The estimate of the co-state equation for the l+1th iteration of the direct yaw moment in the state, is the state quantity after augmentation at time k, ω AFS is the weight coefficient of the middle layer of the active front wheel steering neural network, ω DYC is the weight coefficient of the intermediate layer of the direct yaw moment neural network, ω AFS(l+1) is the weight coefficient of the direct yaw moment for the l+1th iteration, ω DYC(l+1) is the weight coefficient of the direct yaw moment for the l+1th iteration, φ AFS(l+1) is the basis function of the active front wheel steering at the l+1th iteration, φ DYC(l+1) is the basis function of the direct yaw moment for the l+1th iteration.

4. The method according to claim 2, characterized in that: The expression of the target cooperative state in step 2) is as follows: Where D AFS is the weight matrix of active front wheel steering, D DYC is the weight matrix of the direct yaw moment, Q AFS is the tracking error of the active front wheel steering, Q DYC is the tracking error of the direct yaw moment, λ AFS(l+1) (X(k)) is the co-state equation of the l+1th iteration of active front wheel steering under the X(k) state, λ DYC(l+1) (X(k)) is the co-state equation of the l+1th iteration of the direct yaw moment under the X(k) state, γ AFS is the discount factor of the active front steering performance index, γ DYC is the discount factor for the direct yaw moment performance indicator.

5. The method according to claim 2, characterized in that: The specific steps of step 4) are as follows: 41) using the data collected in step 1), the optimization results in step 2), and the vehicle driving condition data in step 3), calculate the state output equation of the closed-loop Nash equilibrium and the state equation of the reference signal system; 42) Calculate the tracking error e of the active front wheel steering AFS (k) and the tracking error of the direct yaw moment e DYC (k); 43) Calculating a cost function to enable the vehicle to track the reference trajectory with minimal energy consumption; 44) Calculate the state space equation of the augmented game system and the performance index function of the augmented game system; 45) Calculate the game output of the weight factor.

6. The method according to claim 5, characterized in that The state output equation of the closed-loop Nash equilibrium in step 41) is: In the formula, x(k+1) is the state quantity at time k+1, x(k) is the state quantity at time k, and u AFS (k) is the input of active front wheel steering at time k, u DYC (k) is the input of the direct yaw moment at time k, A is the coefficient matrix of the state quantity, B AFS is the coefficient matrix of the active front wheel steering input, B DYC is the coefficient matrix of the direct yaw moment input, C AFS is the output matrix of active front wheel steering, C DYC is the output matrix of the direct yaw moment, y AFS (k) is the actual output of active front wheel steering at time k, y DYC (k) is the actual output of the direct yaw moment at time k; The state equation of the reference signal system is: In the formula, is the state of the reference signal system at time k+1, is the state of the reference signal system at time k, r AFS (k) is the expected output of the active front wheel steering reference signal system at time k, r DYC (k) is the expected output of the direct yaw moment reference signal system at time k, F is the system matrix corresponding to the reference signal system, G AFS is the output matrix of the active front wheel steering reference signal system matrix, G DYC is the output matrix of the direct yaw moment reference signal system matrix.

7. The method according to claim 5, characterized in that The tracking error of the active front wheel steering and the direct yaw moment is defined in step 42 as follows: In the formula, e AFS (k) is the tracking error of the active front wheel steering at time k, e DYC (k) is the tracking error of the direct yaw moment at time k.

8. The method according to claim 5, characterized in that In step 45), according to the dynamic non-cooperative game theory, the closed-loop Nash strategy This holds if and only if the following recurrence relation holds: in: In the formula, J is the time k AFS (X(k)) intermediate variables, J is the time k DYC (X(k)) intermediate variables, is the optimal value of active front wheel steering at time k, is the optimal value of the direct yaw moment at time k.

9. The method according to claim 2, characterized in that: The step 5) specifically includes: 51) calculating an estimated input amount of the active front wheel steering for the first iteration and an estimated input amount of the direct yaw moment for the first iteration according to the actual output of the active front wheel steering and the actual output of the direct yaw moment output in step 4); 52) Calculate the target value of the execution network for the lth iteration; 53) Subtract the input estimate of the lth iteration from the target value of the execution network to obtain the error function of the execution network; 54) Adopting the adaptive rule of gradient descent, updating the weights of the execution network, and generating the best active front wheel steering and direct yaw moment control instructions, and completing the control of the vehicle according to the control instructions.

10. The method according to claim 9, characterized in that In step 51), according to the actual output of the active front wheel steering and the actual output of the direct yaw moment output in step 4), the input amount estimation of the lth iteration of the active front wheel steering and the input amount estimation of the lth iteration of the direct yaw moment are calculated: in, is X(k),θ AFS The input estimation of the active front wheel steering in the lth iteration under the state, is X(k),θ DYC The input quantity estimation of the direct yaw moment at the lth iteration in the state, θ AFS is the weight coefficient of the middle layer of the active front wheel steering neural network, θ DYC is the weight coefficient of the intermediate layer of the direct yaw moment neural network, ψ AFS(l) (X(k)) is the basis function of the active front wheel steering at the lth iteration under the X(k) state, ψ DYC(l) (X(k)) is the basis function of the lth iteration of the direct yaw moment in the X(k) state.

Images