Full-vector power chassis automobile driving and braking redundancy cooperative control method

Abstract

The invention discloses a driving and braking redundancy cooperative control method for a full-vector power chassis automobile, relates to the technical field of automobile driving and braking automatic control, and solves the problems that a driving and braking control strategy of an existing full-vector power chassis automobile is not comprehensively considered, and the computing power requirement cannot be well reduced. According to the method, the upper layer calculates the longitudinal moment and the yawing moment required by the vehicle, and the lower layer distributes the longitudinal moment and the yawing moment sent by the upper layer controller in a rule and table look-up combined mode. An intelligent evolution method is used for training an upper-layer actuator neural network and a lower-layer actuator neural network, reinforcement learning control from the vehicle state to instruction output is achieved, the links of signal collection, calculation processing, output execution and the like of a control system are simplified to the maximum degree, particularly the hysteresis effect of the calculation processing link is simplified, and the control efficiency and precision are improved; the economical efficiency, safety and stability of the vehicle in the whole life cycle operation process are guaranteed.

Description

Technical Field

[0001] The present invention relates to the technical field of automobile driving and braking control, and in particular to a method for redundant coordinated control of driving and braking of an automobile with a full vector power chassis. Background Art

[0002] At present, electric vehicles are the hotspot and focus of research in the automotive industry, and full-vector power chassis vehicles are one of the types of electric vehicles that are focused on. They have the advantage of independent driving and braking of four wheels, and can also better control the longitudinal force of each wheel's tires to optimize the vehicle's economy and control yaw stability. And when failure occurs, precise control of each motor and brake can better reduce the harm and loss caused by the failure.

[0003] If a sudden drive or brake failure occurs during vehicle driving and fault-tolerant control is not performed in time, driving safety will be seriously affected. However, the existing economic optimization or stability control strategies of full-vector power chassis vehicles are generally only designed for vehicles in normal driving conditions, and there are few studies on fault-tolerant control strategies for the above-mentioned failure situations. Most of the existing fault-tolerant control strategies for failure situations are only aimed at single drive failure or brake failure, and do not fully integrate the overall coordination of drive and brake, which has limitations and affects the vehicle's operating stability and driving safety.

[0004] For example, the fault-tolerant control strategy in some failure situations is as follows:

[0005] Patent 201910126899.4 discloses a fault-tolerant control strategy for the drive system of a distributed drive electric vehicle. In the upper-level control, the sliding mode variable structure control strategy is used to calculate the desired yaw torque; the lower-level control calculates the distribution of wheel torque through sliding mode variable structure control combined with the desired yaw torque, and redistributes the torque to the wheels that are working normally to ensure the vehicle's driving stability. However, it still has the following shortcomings: both the upper-level calculation and the lower-level distribution use sliding mode algorithms, which require high computing power; the method for expressing torque reconstruction after failure is too simple, and only the control method after a single wheel failure is given, without considering multiple failure situations; only the control method for failure of the drive part is expressed, and there is no expression of the braking system.

[0006] Patent 202010662004.1 discloses a distributed drive electric vehicle drive system control method based on failure status, which classifies and studies the failure conditions of the distributed drive electric vehicle drive system, and further refines the failure control based on the corresponding failure torque redistribution control strategy; but it still has the following shortcomings: it does not mention the strategy expression in the non-failure state, only focuses on the coping strategy in the drive failure state, and only expresses the control method of partial drive failure, without the expression of the braking system.

[0007] Patent 201711285373.8 discloses a hub motor driven automobile electromechanical redundant braking system and control method, which preferentially utilizes the hub motor for regenerative braking, the electromechanical braking system for driving brake compensation and parking brake, and the hydraulic braking system for failure compensation. The three braking systems work together to comprehensively consider road conditions and power battery status, determine the braking mode, solve the total braking torque demand, and reasonably distribute the four-wheel braking torque. The hub motor controller, the electromechanical braking system controller, and the electric mechanical brake booster work together to perform braking control, jointly achieve the braking target and ensure driving stability; but it still has the following shortcomings: only the control method for partial braking failure is expressed, and there is no expression of the drive system; the control strategy under brake failure is not considered; the braking torque distribution form calculated in real time after real-time monitoring of the status has high requirements on computing power.

[0008] It can be seen that the existing control strategy for full-vector power chassis vehicles cannot effectively control when abnormalities occur in the vehicle's driving and braking, and it also does not consider the problem that excessive computing power in the control process may cause control lag. Summary of the invention

[0009] In order to solve the problems existing in the above-mentioned prior art, the present invention provides a redundant coordinated control method for driving and braking of a full-vector power chassis vehicle, which solves the problem that the driving and braking control strategy of the existing full-vector power chassis vehicle is not comprehensive enough and the control strategy cannot be determined with low computing power.

[0010] A fully vector powered chassis vehicle driving and braking redundant coordinated control method, including a hierarchical control method and an intelligent evolution method,

[0011] The hierarchical control method comprises calculating the expected torque according to the vehicle driving information, and then selecting the corresponding torque distribution strategy to distribute the expected torque according to the vehicle state by combining rules and table lookup, wherein the vehicle state comprises a driving state and a braking state, wherein the driving state comprises normal driving and driving failure, the braking state comprises normal braking and braking failure, and the expected torque comprises an expected longitudinal torque and an expected yaw torque;

[0012] The intelligent evolution method includes preparing a data set based on the relevant information of the control process of the hierarchical control method, including vehicle driving information and the corresponding determined torque distribution strategy, to train the neural network model to obtain a strategy generator, and the strategy generator can directly output the expected torque and torque distribution strategy according to the vehicle driving information to control the vehicle;

[0013] The hierarchical control method and the intelligent evolution method are used alternately.

[0014] Moreover, by considering both the driving state and the braking state at the same time, the scope that can be addressed is more comprehensive and the generated strategy is more complete compared to single driving control and braking control.

[0015] Furthermore, the alternating use refers to switching to the intelligent evolution method at the first switching time and switching to the hierarchical control method at the second switching time. The first switching timing refers to the accuracy of the expected torque output by the strategy generator and the torque distribution strategy reaching the expectation, and the second switching timing refers to the vehicle mileage or duration reaching the threshold after adopting the intelligent evolution method.

[0016] First, the hierarchical control method is used to control the vehicle, and data collection training is implemented while ensuring control accuracy. After obtaining enough data, a qualified policy generator is trained and generated. The policy generator can be used directly to replace the complex policy determination process, and the corresponding policy is generated quickly and timely, reducing computing power consumption. When the vehicle mileage reaches a certain value, the state of the vehicle itself changes, and the accuracy of the original policy generator generating the policy will decrease. At this time, the hierarchical control method is re-enabled to control the vehicle, and the data set is collected again to train the policy generator until the accuracy of the policy generated by the policy generator reaches the expected level. Then the policy generator is re-enabled, and this process is repeated, thereby ensuring the stability and safety of the vehicle while minimizing the computing power requirements for generating fault-tolerant policies.

[0017] Furthermore, when braking is normal, a braking torque distribution MAP is used to distribute the desired torque to the brake actuator, and the brake actuator includes an electric brake actuator and a hydraulic brake actuator. The distribution rule of the braking torque distribution MAP is as follows: if the electric brake actuator meets the desired torque, pure electric braking is used; if the electric brake actuator cannot meet the desired torque, the hydraulic brake actuator is used to make up for the shortfall of the electric brake actuator.

[0018] Furthermore, when the brakes fail, a braking torque distribution rule is used to distribute the desired torque to the brake actuator, which includes an electric brake actuator and a hydraulic brake actuator. The braking torque distribution rule is as follows: if the electric brake of the front axle or the rear axle in the electric brake actuator fails, the electric brake and the hydraulic brake actuator corresponding to the axle that has not failed are used for braking; if the electric brakes of both the front axle and the rear axle fail, braking is performed entirely by the hydraulic brake actuator; if the hydraulic brake of the front axle or the rear axle in the hydraulic brake actuator fails, the hydraulic brake and the electric brake actuator corresponding to the axle that has not failed are used for braking; if the hydraulic brakes of both the front axle and the rear axle fail, braking is performed entirely by the electric brake actuator; if the electric brake actuator is insufficient to provide the desired torque, the vehicle speed is reduced to reduce the braking demand torque and the vehicle stops.

[0019] Furthermore, when the driving is normal, the driving torque distribution MAP is used to distribute the desired torque, and the driving torque distribution MAP is formed as follows:

[0020] Define the torque distribution coefficient λ:

[0021]

[0022] Where, T d is the desired torque, T fl , T fr is the torque distributed to the left front wheel and the right front wheel, T rl , T rr is the torque distributed to the left rear wheel and the right rear wheel, λ is the torque distribution coefficient of the front and rear axles;

[0023] When λ=0, it indicates rear-wheel drive, and when λ=1, it indicates front-wheel drive. The total driving efficiency is shown in the following formula:

[0024]

[0025] Where η fl (T fl , n * ) is the left front wheel torque T fl At speed n * The front wheel efficiency under fr (T fr , n * ) is the torque T fr At speed n * The right front wheel efficiency under rl (T rl , n * ) is the torque T rl At speed n * The efficiency of the left rear wheel under rr (T rr , n * ) is the torque T rr At speed n * The right rear wheel efficiency under the condition of η is the total driving efficiency of the vehicle, T d is the driver's desired torque, η fl , η fr , η rl , η rr They are the left front, right front, left rear and right rear wheel motor speeds respectively;

[0026] According to the above formula, the traversal cycle calculation is performed to obtain the torque distribution coefficient with the best efficiency between different total expected torques and motor speeds, which is λ, that is, the expected torque T is obtained. d , motor speed as input, and torque distribution coefficient λ as output drive torque distribution MAP.

[0027] Furthermore, when the drive fails, the failure position is first determined and then processed according to established rules. The failure positions of the drive failure include single motor failure, coaxial dual motor failure, same-side dual motor failure, different-axis different-side dual motor failure and three-motor failure. If the failure position is single motor failure or coaxial dual motor failure, the motor corresponding to the shaft that has not failed will provide the expected torque. If the failure position is different-axis different-side dual motor failure, the non-failed motor will provide the expected torque. If the failure position is same-side dual motor failure or three-motor failure, the non-failed motor will provide the expected torque. At the same time, the output torque of a single non-failed motor is kept less than a calibrated value to ensure that the vehicle travels at a low speed and avoid the imbalance of the left and right torques affecting safety.

[0028] Further, the strategy generator includes a torque generation module and a torque distribution module, the torque generation module is used to directly output the desired torque, and the torque distribution module is used to directly determine the torque distribution strategy according to the desired torque and the vehicle state;

[0029] For the torque generation module, its state space and execution actions can be defined as:

[0030]

[0031] Where α is the accelerator pedal signal, β is the brake pedal signal, and F x is the desired longitudinal moment, ΔM is the desired yaw moment; is the longitudinal speed; V y is the lateral vehicle speed; z is the yaw angular velocity;

[0032] For the torque distribution module, its state space and execution actions can be defined as:

[0033]

[0034] In the formula, E Tij(i＝1，2；j＝1，2) For the four motor states, E Bij(i＝1，2；j＝1，2) There are four hydraulic brake states, T Tij(i＝1，2；j＝1，2) is the torque command output to the four motors, T Bij(i＝1，2；j＝1，2) It is the braking torque command output to the four hydraulic brakes.

[0035] Furthermore, in the process of calculating the expected longitudinal torque, the expected speed is used as the control target, and in the process of calculating the expected yaw torque, the yaw angular velocity and the center of mass deflection angle are used as control targets. The expected yaw moment is determined by the vehicle body posture. When the vehicle is not unstable, the expected yaw moment is 0, and the expected torque is the expected longitudinal torque. When the vehicle has a tendency to become unstable, the expected torque includes both the expected driving torque and the expected yaw moment.

[0036] Furthermore, the vehicle driving information includes the vehicle operating status fed back by the vehicle posture sensor and the driver operation signal, that is, the high-level intelligent driving instruction includes an accelerator pedal signal and a brake pedal signal.

[0037] Furthermore, the hierarchical control method includes upper-level control and lower-level control, wherein the upper-level control includes calculating the desired torque based on vehicle driving information; the lower-level control includes determining the state of a drive actuator or a brake actuator, and then selecting a corresponding torque distribution strategy to distribute the desired torque based on the state by combining rules with table lookup.

[0038] The beneficial effects of the present invention include:

[0039] (1) The coordinated consideration of the drive system actuator and brake system actuator of the full vector power chassis vehicle is more complete and comprehensive than the existing rule-based drive and brake fault handling methods or separate drive stability control strategies.

[0040] (2) The vehicle driving economy, vehicle stability, braking recovery energy utilization efficiency, and braking smoothness can be comprehensively considered under the premise of integrating drive and brake failure tolerance control.

[0041] (3) The control strategy is implemented by combining rules with MAP lookup, which saves computing power while ensuring the economy, safety, and stability of the vehicle throughout its life cycle.

[0042] (4) The intelligent evolution method is used to train the upper and lower actuator neural networks, realizing reinforcement learning control based on vehicle status to command output, which greatly simplifies the control system's "signal acquisition-computation processing-output execution" and other links, especially the hysteresis effect of the calculation processing link, and improves control efficiency and accuracy; BRIEF DESCRIPTION OF THE DRAWINGS

[0043] Figure 1 The present invention is a flow chart of a method for redundant coordinated control of driving and braking of a full vector power chassis vehicle. DETAILED DESCRIPTION

[0044] In order to make the purpose, technical scheme and advantages of the embodiments of the present application clearer, the technical scheme in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all of the embodiments. Therefore, the following detailed description of the embodiments of the present application provided in the drawings is not intended to limit the scope of the application for protection, but merely represents the selected embodiments of the present application. Based on the embodiments of the present application, all other embodiments obtained by those skilled in the art without making creative work belong to the scope of protection of the present application.

[0045] A fully vector powered chassis vehicle driving and braking redundant coordinated control method, including a hierarchical control method and an intelligent evolution method,

[0046] The hierarchical control method comprises calculating the expected torque according to the vehicle driving information, and then selecting the corresponding torque distribution strategy to distribute the expected torque according to the vehicle state, wherein the vehicle state comprises a driving state and a braking state, wherein the driving state comprises normal driving and driving failure, wherein the braking state comprises normal braking and braking failure, and wherein the expected torque comprises an expected longitudinal torque and an expected yaw torque;

[0047] The intelligent evolution method includes preparing a data set based on relevant information of the control process of the hierarchical control method, including vehicle driving information and a corresponding determined torque distribution strategy, and training a neural network model to obtain a strategy generator. The strategy generator can directly output a torque distribution strategy to control the vehicle based on the vehicle driving information; the hierarchical control method and the intelligent evolution method are used alternately.

[0048] Taking both driving and braking states into consideration at the same time, compared with single driving control and braking control, the scope that can be addressed is more comprehensive and the generated strategy is more complete.

[0049] Specifically, at a specific moment, the vehicle can only be in a specific state between the driving state and the braking state. When the vehicle is in the driving state, the expected torque is the expected driving torque and the expected yaw torque. When the vehicle is in the braking state, the expected torque is the expected braking torque and the expected yaw torque.

[0050] When in the driving state, the desired torque is calculated by the driving motion tracking controller and distributed by the driving distribution controller. When in the braking state, the desired torque is calculated by the braking motion tracking controller and distributed by the braking distribution controller.

[0051] When in driving state, the drive motion tracking controller will select a strategy based on the failure state of the drive motor. When there is no motor failure, the front and rear axle torque distribution adopts the form of MAP lookup to improve the vehicle driving economy. When a motor failure occurs, torque distribution and torque reconstruction are performed according to established rules to ensure vehicle stability and power.

[0052] When in the braking state, the brake motion tracking controller will select a strategy based on the failure state of the brake actuator (motor brake and hydraulic brake). When there is no brake actuator failure, the electro-hydraulic brake distribution ratio adopts the form of MAP lookup to improve the vehicle energy recovery efficiency. When the brake actuator fails, the braking torque is distributed and reconstructed according to the established rules to ensure vehicle stability and braking performance. After the relevant MAP is written to the controller for the first time, DQN learning training and periodic iteration are carried out continuously during operation.

[0053] In another embodiment, the hierarchical control method and the intelligent evolution method are used alternately; the alternating use refers to switching to the intelligent evolution method at a first switching time and switching to the hierarchical control method at a second switching time, the first switching time refers to the strategy generator outputting the expected torque and the accuracy of the torque distribution strategy reaching the expectation, and the second switching time refers to the vehicle mileage or duration reaching a threshold after adopting the intelligent evolution method.

[0054] First, the hierarchical control method is used to control the vehicle, and data collection training is implemented while ensuring control accuracy. After obtaining enough data, a qualified policy generator is trained and generated. The policy generator can be used directly to replace the complex policy determination process, and the corresponding policy is generated quickly and timely, reducing computing power consumption. When the vehicle mileage reaches a certain value, the state of the vehicle itself changes, and the accuracy of the original policy generator generating the policy will decrease. At this time, the hierarchical control method is re-enabled to control the vehicle, and the data set is collected again to train the policy generator until the accuracy of the policy generated by the policy generator reaches the expected level. Then the policy generator is re-enabled, and this process is repeated, thereby ensuring the stability and safety of the vehicle while minimizing the computing power requirements for generating fault-tolerant policies.

[0055] Specifically, since the state changes of the vehicle during its entire life cycle will affect the control accuracy of the reinforcement learning network, the vehicle mileage interval (not less than 10,000 kilometers) and the update time interval (not less than half a year) are used as the basis conditions. After the judgment conditions are met, the hierarchical control method is re-adopted to perform the "signal acquisition-computation processing-output execution" link to take over the vehicle control. After a period of continuous control with sufficient data, the intelligent evolution method is used to retrain the strategy generator. After the training is completed, it is replaced using the same method to achieve evolutionary iteration.

[0056] In another embodiment, when braking is normal, a braking torque distribution MAP is used to distribute the desired torque to the brake actuator, and the brake actuator includes an electric brake actuator and a hydraulic brake actuator. The distribution rule of the braking torque distribution MAP is: if the electric brake actuator meets the desired torque, pure electric braking is used; if the electric brake actuator cannot meet the desired torque, the hydraulic brake actuator is used to make up for the shortfall of the electric brake actuator.

[0057] Specifically, in the braking execution part, under fault-free conditions, the braking torque distribution between the brake actuators is determined by the braking torque distribution MAP (MAP_B for short), which is a distribution MAP based on the economic concept. At this time, in order to maximize the energy recovery and energy saving effect, if the electric braking torque meets the braking torque requirement, pure electric braking is performed; if the motor can participate in braking, but the electric braking torque cannot meet the braking torque requirement of the vehicle, the hydraulic braking system is activated, and the hydraulic braking torque is used to make up for the insufficient braking torque required by the vehicle. Based on this concept, the total braking torque distribution coefficient can be As a parameter of MAP_B output. Defines the total motor braking torque T BM , hydraulic brake total torque T BH and the expected longitudinal moment T b Satisfies the following relationship

[0058]

[0059] On this basis, to ensure safety and prevent the motor braking command from exceeding the threshold, the electric braking torque output by each motor is T BM / 4 is no longer adjusted, each hydraulic brake is at output T BH On the basis of the / 4 braking torque, an additional yaw torque ΔM is provided by adjusting the left and right side braking torques.

[0060] In another embodiment, when the brakes fail, a braking torque distribution rule is used to distribute the desired torque to the brake actuator, which includes an electric brake actuator and a hydraulic brake actuator. The braking torque distribution rule is as follows: if the electric brake of the front axle or the rear axle in the electric brake actuator fails, the electric brake and the hydraulic brake actuator corresponding to the axle that has not failed are used for braking; if the electric brakes of both the front axle and the rear axle fail, braking is performed entirely by the hydraulic brake actuator; if the hydraulic brake of the front axle or the rear axle in the hydraulic brake actuator fails, the hydraulic brake and the electric brake actuator corresponding to the axle that has not failed are used for braking; if the hydraulic brakes of both the front axle and the rear axle fail, braking is performed entirely by the electric brake actuator; if the electric brake actuator is insufficient to provide the desired torque, the vehicle speed is reduced to reduce the braking demand torque and the vehicle stops.

[0061] Specifically, electric brake failure mainly refers to the motor being unable to output torque commands, the traction motor speed being too low (brake energy recovery has limitations on the motor speed. From the traction motor power generation efficiency characteristics, it can be seen that when the traction motor is involved in braking, if the speed is low, both the motor power generation efficiency and the motor control accuracy are not conducive to motor braking. At the same time, the large torque at low speed will cause the motor to generate more heat; if the motor speed is too low, the motor efficiency is too low, and the high heat will affect the motor life. Therefore, when the motor speed is lower than a certain value, it is considered as an electric brake failure), and the traction motor temperature is too high (high heat will affect the motor life, so when the motor temperature is too high, it is considered as an electric brake failure).

[0062] When there is a fault, the four motors are divided into two groups for motor braking failure: the two motors on the front axle are grouped into the front axle motor group, and the two motors on the rear axle are grouped into the rear axle motor group. To ensure stability during braking, it is not allowed for one motor in the same group to brake while the other motor does not brake. Only both motors can brake or neither motor can brake.

[0063] Take the front axle as an example. If one of the motors on the front axle cannot participate in braking due to its speed, temperature, or failure, then both motors on the front axle will not brake. At this moment, to ensure safety, the rear axle motor and hydraulic brake will brake normally, and the braking torque that the two motors on the front axle cannot provide will be compensated by the front axle hydraulic brake. The rear axle motor failure is handled in the same way. If both the front and rear axles have motors that cannot participate in braking, then all braking torques and additional yaw torques will be provided by the hydraulic brake.

[0064] Hydraulic brake failure mainly refers to the inability of the hydraulic brake to complete the braking action.

[0065] Similarly, the two hydraulic brakes on the front axle are grouped into the front axle hydraulic brake group, and the two hydraulic brakes on the rear axle are grouped into the rear axle hydraulic brake group. To ensure stability during braking, it is not allowed for one hydraulic brake in the same group to brake while the other hydraulic brake is not braked. Only both brakes or neither brake is allowed.

[0066] Taking the front axle as an example, if one of the hydraulic brakes on the front axle cannot participate in braking, then both hydraulic brakes on the front axle will not brake. At this moment, to ensure safety, the front axle motor, rear axle motor and rear axle hydraulic brake will brake normally, and the braking torque that the two hydraulic brakes on the front axle cannot provide will be compensated by the rear axle hydraulic brake. The fault handling method for the rear axle hydraulic brake is similar. If both the front and rear axles have hydraulic brake failures, since the motor brake cannot provide a large braking torque, in order to ensure safety when this state occurs, the driver should be reminded immediately and the vehicle should be slowed down and stopped in time.

[0067] In another embodiment, when the driving is normal, the driving torque distribution MAP (abbreviated as MAP_D) is used to distribute the desired torque, and the formation process of the MAP_D is as follows:

[0068] Define the torque distribution coefficient λ:

[0069]

[0070] Where, T d is the expected longitudinal moment, T fl 、T fr is the torque distributed to the left front wheel and the right front wheel, T rl 、T rr is the torque distributed to the left rear wheel and the right rear wheel, λ is the torque distribution coefficient of the front and rear axles;

[0071] When λ=0, it indicates rear-wheel drive, and when λ=1, it indicates front-wheel drive. The total driving efficiency is shown in the following formula:

[0072]

[0073] Where η fl (T fl , n * ) is the left front wheel torque T fl , speed n * The front wheel efficiency under fr (T fr , n * ) is the torque T fr , speed n * The right front wheel efficiency under rl (T rl , n * ) is the torque T rl , speed n * The efficiency of the left rear wheel under rr (T rr , n * ) is the torque T rr , speed n * The right rear wheel efficiency under the condition of η is the total driving efficiency of the four driving motors, T d is the driver's desired torque, η fl , η fr , η rl , η rr They are the left front, right front, left rear and right rear wheel motor speeds respectively;

[0074] According to the above formula, the traversal cycle calculation is performed to obtain the torque distribution coefficient with the best efficiency between different total expected torques and motor speeds, which is λ, that is, the expected torque T is obtained. d , motor speed as input, and torque distribution coefficient λ as output MAP_D.

[0075] Specifically, when the driving is normal, that is, when there is no fault, the corresponding driving torque of each wheel satisfies the following conditions:

[0076]

[0077] Among them, B is the track width of the front and rear axles (generally considered equal), and r is the tire radius.

[0078] In another embodiment, when the drive fails, the failure position is first determined and then processed according to established rules. The failure positions of the drive failure include single motor failure, coaxial dual motor failure, same-side dual motor failure, different-axis different-side dual motor failure and three-motor failure. If the failure position is single motor failure or coaxial dual motor failure, the motor corresponding to the shaft that has not failed will provide the expected torque. If the failure position is different-axis different-side dual motor failure, the non-failed motor will provide the expected torque. If the failure position is same-side dual motor failure or three-motor failure, the non-failed motor will provide the expected torque. At the same time, the output torque of a single non-failed motor is kept less than a calibrated value to ensure that the vehicle travels at a low speed and avoid the imbalance of the left and right torques affecting safety.

[0079] Specifically, the processing methods corresponding to each failure position are exemplified as follows:

[0080]

[0081]

[0082] In another embodiment, the strategy generator includes a torque generation module and a torque distribution module, wherein the torque generation module is used to directly output the desired torque, and the torque distribution module is used to directly determine the torque distribution strategy according to the desired torque and the vehicle state;

[0083] For the torque generation module, its state space and execution actions can be defined as:

[0084]

[0085] Where α is the driver's accelerator pedal signal (intelligent driving system acceleration command), β is the driver's brake pedal signal (intelligent driving system brake command), and F x is the desired longitudinal moment, ΔM is the desired yaw moment; is the longitudinal speed; V y is the lateral vehicle speed; z is the yaw rate

[0086] For the torque distribution module, its state space and execution actions can be defined as:

[0087]

[0088] In the formula, ETij(i＝1，2；j＝1，2) is the state of the four motors, E Bij(i＝1，2；j＝1，2) There are four hydraulic brake states, T Tij(i＝1，2；j＝1，2) is the torque command output to the four motors, T Bij(i＝1，2；j＝1，2) To output the braking torque instructions to the four hydraulic brakes, specifically, when in the driving state, the four motors drive the corresponding wheels respectively, and when in the braking state, the four motors act as electric brakes to brake the corresponding wheels respectively.

[0089] In this embodiment, the reinforcement learning method adopts the DQN reinforcement learning method, and the specific contents are as follows:

[0090] The intelligent agent of reinforcement learning interacts with the environment by selecting actions with the maximum future feedback value, and calculates the action Q value by defining the optimal action selection function Q(s, a);

[0091] Q(s,a)=max π E[R t / s t =s,a t =a / π]

[0092] Among them, s is the state, a is the action performed in this state; π is the mapping between action and state; s t is the state at time step t, a t For t Actions performed in the state; R t The feedback value obtained by executing action a when the state is s is as follows:

[0093]

[0094] Among them, t is the time step; T is the total time step when the terminal state is reached; r t′ is the feedback value of the t′th time step; γ represents the decay rate of the feedback value at each time step;

[0095] Deep learning uses the samples provided by reinforcement learning to obtain the Q values ​​of all actions in this state, and calculates the loss function L with the correct Q values ​​provided by reinforcement learning. i (θ i ) as shown in formula (5); further, the weight θ of the neural network approximator is updated by using the stochastic gradient descent method on the loss function, so as to achieve the purpose of optimizing the Q neural network;

[0096] L i (θ i )=E s，a～ρ(·) [(y i -Q(s, a; θ i )) 2 ]

[0097] Among them, Q(s, a; θ i ) is the estimated value of Q(s, a); i is the number of iterations; s is the current state; a is the currently selected action; ρ(s, a) is the probability distribution between state s and action a; y i is Q(s, a) obtained according to the Bellman equation, which can be expressed as formula (6);

[0098] y i =E s′ [r+γmax a′ Q i (s′, a′; θ i-1 ) / s,a]

[0099] Among them, s′ is the next state; a′ is the next action.

[0100] In another embodiment, the expected speed is used as the control target in the calculation process of the expected longitudinal torque, and the yaw angular velocity and the center of mass deflection angle are used as the control targets in the calculation process of the expected yaw moment. The expected yaw moment is determined by the vehicle body posture. When the vehicle is not unstable, the expected yaw moment is 0, and the expected torque is the expected longitudinal torque. When the vehicle has a tendency to become unstable, the expected torque includes both the expected driving torque and the expected yaw moment.

[0101] In another embodiment, the vehicle driving information includes the vehicle operating status fed back by the vehicle posture sensor and the driver operation signal, that is, the high-order intelligent driving instructions include the driver's accelerator pedal signal (intelligent driving system acceleration instruction) and the driver's brake pedal signal (intelligent driving system braking instruction).

[0102] In another embodiment, the hierarchical control method includes upper-level control and lower-level control, wherein the upper-level control includes calculating the desired torque based on vehicle driving information; the lower-level control includes determining the state of a drive actuator or a brake actuator, and then selecting a corresponding torque distribution strategy to distribute the desired torque based on the state.

[0103] The above-mentioned embodiments only express the specific implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the protection scope of the present application. It should be pointed out that, for ordinary technicians in this field, several variations and improvements can be made without departing from the technical solution concept of the present application, and these all belong to the protection scope of the present application.

Claims

1. A method for redundant coordinated control of driving and braking of a full vector power chassis vehicle, characterized in that: Including hierarchical control methods and intelligent evolution methods, The hierarchical control method comprises calculating the expected torque according to the vehicle driving information, and then selecting the corresponding torque distribution strategy to distribute the expected torque according to the vehicle state by combining rules and table lookup, wherein the vehicle state comprises a driving state and a braking state, wherein the driving state comprises normal driving and driving failure, the braking state comprises normal braking and braking failure, and the expected torque comprises an expected longitudinal torque and an expected yaw torque; The intelligent evolution method includes preparing a data set based on the relevant information of the control process of the hierarchical control method, including vehicle driving information and the corresponding determined torque distribution strategy, to train the neural network model to obtain a strategy generator, and the strategy generator can directly output the expected torque and torque distribution strategy according to the vehicle driving information to control the vehicle; The hierarchical control method and the intelligent evolution method are used alternately.

2. A method for redundant coordinated control of driving and braking of a full vector power chassis vehicle according to claim 1, characterized in that: The alternating use refers to switching to the intelligent evolution method at the first switching time and switching to the hierarchical control method at the second switching time. The first switching timing refers to the accuracy of the expected torque output by the strategy generator and the torque distribution strategy reaching the expectation, and the second switching timing refers to the vehicle mileage or duration reaching the threshold after using the intelligent evolution method.

3. The method for redundant coordinated control of driving and braking of a full vector power chassis vehicle according to claim 1 is characterized in that: When braking is normal, the desired torque is distributed using the braking torque distribution MAP. The braking actuator includes an electric braking actuator and a hydraulic braking actuator. The distribution rule of the braking torque distribution MAP is as follows: if the electric braking actuator meets the desired torque, pure electric braking is used. If the electric braking actuator cannot meet the desired torque, the hydraulic braking actuator is used to make up for the shortfall of the electric braking actuator.

4. The method for controlling driving and braking redundancy of a full vector power chassis vehicle according to claim 1 is characterized in that: When the brakes fail, a braking torque distribution rule is used to distribute the desired torque to the brake actuator, which includes an electric brake actuator and a hydraulic brake actuator. The braking torque distribution rule is as follows: if the electric brake of the front axle or the rear axle in the electric brake actuator fails, the electric brake and the hydraulic brake actuator corresponding to the axle that has not failed are used for braking; if the electric brakes of both the front axle and the rear axle fail, braking is performed entirely by the hydraulic brake actuator; if the hydraulic brake of the front axle or the rear axle in the hydraulic brake actuator fails, the hydraulic brake and the electric brake actuator corresponding to the axle that has not failed are used for braking; if the hydraulic brakes of both the front axle and the rear axle fail, braking is performed entirely by the electric brake actuator; if the electric brake actuator is insufficient to provide the desired torque, the vehicle speed is reduced to reduce the braking demand torque and the vehicle stops.

5. The method for redundant coordinated control of driving and braking of a full vector power chassis vehicle according to claim 1 is characterized in that: When the driving is normal, the desired torque is distributed to the brake actuator using the driving torque distribution MAP, and the driving torque distribution MAP is formed as follows: Define the torque distribution coefficient λ: Where, T d is the desired torque, T fl , T fr is the torque distributed to the left front wheel and the right front wheel, T rl , T rr is the torque distributed to the left rear wheel and the right rear wheel, λ is the torque distribution coefficient of the front and rear axles; When λ=0, it indicates rear-wheel drive, and when λ=1, it indicates front-wheel drive. The total driving efficiency is shown in the following formula: Where η fl (T fl , n * ) is the left front wheel torque T fl At speed n * The front wheel efficiency under fr (T fr , n * ) is the torque T fr At speed n * The right front wheel efficiency under rl (T rl , n * ) is the torque T rl At speed n * The efficiency of the left rear wheel under rr (T rr , n * ) is the torque T rr At speed n * The right rear wheel efficiency under the condition of η is the total driving efficiency of the vehicle, T d is the driver's desired torque, η fl , η fr , η rl , η rr They are the left front, right front, left rear and right rear wheel motor speeds respectively; According to the above formula, the traversal cycle calculation is performed to obtain the torque distribution coefficient λ with the best efficiency between different total expected torques and motor speeds, that is, the expected torque T is obtained. d , motor speed as input, and torque distribution coefficient λ as output drive torque distribution MAP.

6. The method for redundant coordinated control of driving and braking of a full vector power chassis vehicle according to claim 1 is characterized in that: When the drive fails, the failure position is first determined and then processed according to the established rules. The failure positions of the drive failure include single motor failure, coaxial dual motor failure, same-side dual motor failure, different-axis different-side dual motor failure and three-motor failure. If the failure position is single motor failure or coaxial dual motor failure, the motor corresponding to the axis that has not failed will provide the expected torque. If the failure position is different-axis different-side dual motor failure, the motor that has not failed will provide the expected torque. If the failure position is same-side dual motor failure or three-motor failure, the motor that has not failed will provide the expected torque, and at the same time, the output torque of a single non-failed motor will be kept less than the calibration value.

7. The method for controlling driving and braking redundancy of a full vector power chassis vehicle according to claim 1 is characterized in that: The strategy generator includes a torque generation module and a torque distribution module, wherein the torque generation module is used to directly output the desired torque, and the torque distribution module is used to directly determine the torque distribution strategy according to the desired torque and the vehicle state; For the torque generation module, its state space and execution actions can be defined as: Where α is the accelerator pedal signal, β is the brake pedal signal, and F x is the desired longitudinal moment, ΔM is the desired yaw moment; is the longitudinal vehicle speed; is the lateral vehicle speed; z is the yaw angular velocity; For the torque distribution module, its state space and execution actions can be defined as: In the formula, ET ij(i＝1,2；j＝1,2) For the four motor states, E Bij(i＝1,2；j＝1,2) There are four hydraulic brake states, T Tij(i＝1,2；j＝1,2) is the torque command output to the four motors, T Bij(i＝1,2；j＝1,2) It is the braking torque command output to the four hydraulic brakes.

8. A method for redundant coordinated control of driving and braking of a full vector power chassis vehicle according to any one of claims 1 to 7, characterized in that: In the process of calculating the expected longitudinal moment, the expected speed is used as the control target. In the process of calculating the expected yaw moment, the yaw angular velocity and the center of mass deflection angle are used as the control targets. The expected yaw moment is determined by the vehicle body posture. When the vehicle is not unstable, the expected yaw moment is 0. At this time, the expected moment is the expected longitudinal moment. When the vehicle has a tendency to become unstable, the desired torque includes both the desired driving torque and the desired yaw torque.

9. A method for redundant coordinated control of driving and braking of a full vector power chassis vehicle according to any one of claims 1 to 7, characterized in that: The vehicle driving information includes the vehicle operating status fed back by the vehicle posture sensor and the driver operation signal, that is, the high-level intelligent driving instruction includes the accelerator pedal signal and the brake pedal signal.

10. A method for redundant coordinated control of driving and braking of a full vector power chassis vehicle according to any one of claims 1 to 7, characterized in that: The hierarchical control method includes upper-level control and lower-level control. The upper-level control includes calculating the expected torque based on vehicle driving information; the lower-level control includes judging the state of a driving actuator or a braking actuator, and then selecting a corresponding torque distribution strategy to distribute the expected torque based on the state by combining rules with table lookup.

Images