An event-triggered-based drive-by-wire corner tracking control method, storage medium and device

Abstract

The application discloses a kind of based on event trigger's drive-by-wire corner tracking control method, first introduce radial basis neural network algorithm simulation model in nonlinear variable, establish drive-by-wire steering system actuator dynamics model.Then, the dynamics model established is designed to the corner tracking controller, and event trigger condition is designed, so that the time output of steering execution motor produces interval, reduces the occupation for communication resources.Next, specific control rate parameter is selected, the weight of neural network in control rate and Nussbaum function are updated, combined with the sliding mode control rate, the torque control rate of drive-by-wire steering system actuator is obtained, and steering motor torque is calculated.Finally, steering wheel is turned according to expected path by controlling steering execution motor torque, and the neural network used has the ability to fit nonlinear torque disturbance, and has certain fault tolerance.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of automobile steering, and more particularly to an event-triggered steer-by-wire corner tracking control method. In addition, the present application also relates to a storage medium capable of providing the above tracking control method, and an apparatus capable of implementing the above tracking control method. BACKGROUND

[0002] The automobile steering system is an important component of an automobile, which has experienced multiple stages such as traditional mechanical transmission steering, electric power steering, and steer-by-wire. Nowadays, the traditional mechanical transmission steering has been eliminated, and most automobiles adopt the electric power steering system. The electric power steering system relies on an electric motor to provide auxiliary steering power, which can reduce the driving and steering pressure for the driver, and on some more advanced automobiles, the degree of electric assistance can be adjusted according to the driving mode to improve the driving pleasure. In the future, more and more new energy vehicles will be introduced, and the automobile steering system will inevitably develop to the stage of steer-by-wire to further reduce the number of automobile parts and reduce the cost of the automobile. The steer-by-wire system removes the steering column and uses pure electric signals to complete the steering operation, which has the advantages of small size, flexible arrangement, optimized driving experience, reduced driver fatigue, and guaranteed vehicle safety.

[0003] The steer-by-wire system commonly used in the prior art processes the equivalent moment of inertia and equivalent damping coefficient of the uncertain parameters by establishing a steer-by-wire system model, obtains the lumped uncertainty, designs an extended state observer tracking control algorithm, and redesigns the controller to make the steering wheel steering angle track the steering wheel command given by the driver. The corner tracking error and the fractional order sliding mode surface are defined to obtain a fractional order sliding mode controller based on the extended state observer. By constructing a Lypunov function, the fractional order sliding mode controller satisfies the asymptotic stability condition of the steer-by-wire system, so that it still has good control effect when the system parameters are perturbed or there is external disturbance. However, this control method can handle the uncertainty in the steer-by-wire system, but does not have fault tolerance capability, and the sensors are prone to failure due to various external disturbances. Moreover, the torque output mechanism based on time triggering occupies a lot of communication resources, and frequent communication exchanges inevitably cause data transmission packet loss, out-of-order, and delay phenomena, which have a large safety hazard. SUMMARY

[0004] The present application aims at overcoming the defects of the existing tracking control method of steer-by-wire system, which has no fault tolerance and occupies more communication resources, and provides a steer-by-wire angle tracking control method based on event triggering, which can ensure the basic performance of the steer-by-wire system angle tracking control under the condition of occupying less communication resources, has fast dynamic response speed and high tracking accuracy, and has certain fault tolerance, which can effectively handle the working condition of steering motor failure.

[0005] To solve the above technical problems, the technical scheme adopted by the present application is to provide a steer-by-wire angle tracking control method based on event triggering, which comprises the following steps:

[0006] Step S1, introducing a radial basis neural network algorithm to simulate the nonlinear variables in the model, determining the input, state and output of the control system, and establishing a steer-by-wire system actuator dynamics model;

[0007] Step S2, designing the angle tracking controller for the dynamics model established in step S1, introducing a dynamic sliding mode surface method, a Nussbaum function and an event triggering condition, and using a backstepping method to derive a sliding mode control rate, so that the system converges in a limited time and meets the Lyapunov stability criterion;

[0008] Step S3, selecting specific control rate parameters, updating the weights of the neural network and the Nussbaum function in the control rate of step S2, combining the sliding mode control rate to obtain the torque control rate of the steer-by-wire system actuator, and calculating the steering motor torque;

[0009] Step S4, realizing the steering of the steering wheel according to the expected path by controlling the torque of the steering motor.

[0010] The working principle of the application is: the rotation angles of the steering wheel and the automobile steering wheel are monitored by a steering sensor, and the two monitoring results are synchronized into a radial basis neural network to establish a dynamics model of a steer-by-wire system actuator. Compared with the prior art, the application introduces a Nussbaum function to process the failure of the execution motor. The Nussbaum function can generally be used to solve the stabilization problem of some systems with uncertain control coefficients. Specifically, when the execution motor fails or has a bias type failure, the steering sensor can monitor the rotation angle error of the steering wheel and the automobile steering wheel. When the rotation angle error is greater than a set value, the steering execution motor can powerfully compensate the steering wheel, so that the rotation angle of the steering wheel corresponds to the rotation angle of the steering wheel controlled by the driver, and the steering precision effect is achieved. Of course, in order to consider the frequent failure problems in the system, a filter needs to be additionally introduced to facilitate the design of the rotation angle tracking controller. At the same time, in order to save communication resources and avoid occupying too much communication resources to reduce the rotation angle tracking precision, the application also designs an event triggering condition, so that the time interval of the steering execution motor output is generated, and only when the output torque of the steering execution motor meets the set condition, the output torque will be updated, otherwise the torque output will be according to the torque output at the last time. In this way, the torque output frequency of the steering execution motor can be effectively reduced. Unlike the traditional time triggering mechanism, the output torque of the steering execution motor under the event triggering mechanism is discontinuous, thereby achieving the purpose of saving the occupancy rate of communication resources.

[0011] Further, the steer-by-wire system actuator dynamics model of step S1 can be expressed by the following formula:

[0012]

[0013] In the formula, J eq represents the total rotation torque, which can be equal to wherein J f represents the steering wheel kingpin rotation inertia, K c represents the transmission ratio between the motor rotation angle and the steering wheel rotation angle, J m2 represents the execution motor rotation inertia;

[0014] B eq represents the total viscous damping coefficient, which can be equal to wherein B f represents the damping coefficient of the steering wheel kingpin, K c represents the transmission ratio between the motor rotation angle and the steering wheel rotation angle, B m2 represents the execution motor damping coefficient;

[0015] represents the total torque disturbance, and f ​respectively, which can be expressed as τ e +τ f , τ e is the self-aligning moment of the tire, τ f is the equivalent friction moment on the kingpin; T m represents the output of the motor.

[0016] Further, in step S1, the total moment disturbance is approximated by the radial basis neural network, using the following formula:

[0017]

[0018] In the formula, represents the input node of the radial basis neural network, represents the output weight of each hidden layer, and the superscript T represents the transpose, and the vector φ(x) ∈ R N is a Gaussian function, which is defined as:

[0019]

[0020] In the formula, g i and σ i are the center value and width of the Nth neural network node, respectively.

[0021] Further, in step S1, the steering actuator torque is defined as the system input, the steering wheel angle and speed are defined as the system state, and the actual steering angle is defined as the system output, while considering the actuator fault in the system output, to obtain the state space expression of the steering actuator:

[0022]

[0023] In the formula, define and G(t) = β(t-T)g(x), g(x) is a fault that occurs at β(t-T), and define T n as the time of fault occurrence, when t < T n , G(t) = 1, when t ≥ T n , G(t) = 1 + g( x ), and u as the control input value is ε1 represents the error of neural network estimation, and d(t) represents unknown external disturbance;

[0024] x1 represents the actual rotation angle of the steering wheel, x2 represents the angular velocity of the steering wheel, and y represents the output of the motor.

[0025] Further, the dynamic sliding mode surface method described in step S2 includes the following steps:

[0026] Step S21, calculating the virtual control inverse using an integral filter, which is designed as follows:

[0027]

[0028] where τ is a time constant, denotes the time derivative of the output, denotes the output of a first order low pass filter a, which is calculated as follows:

[0029]

[0030] where s1 is the tracking error, which denotes the error between the steering wheel angle and the desired angle, s2 is the derivative of the tracking error, which denotes the error of the steering wheel angle velocity, and v denotes the output error of the filter, all of which can be monitored by the steering sensor;

[0031] Step S22, designing the control law of the steering execution motor using the filter:

[0032]

[0033] where p(t) denotes the torque output value of the steering execution motor, N(ξ) is the Nussbaum function, c2, and ε is a positive adjustable parameter, which respectively denotes the disturbance rejection ability of the sliding mode surface, the maximum error of the model, and the speed of the sliding mode surface tending to the limit, denotes the actual estimation of the total disturbance torque by the neural network;

[0034] Step S23, setting the following event trigger condition for the torque output value, so that the output torque of the motor is only updated when the condition is met, otherwise the torque output of the last time will be used:

[0035]

[0036] where u(t) denotes the actual output torque of the steering execution motor, p(t k ) denotes the output torque of the steering motor at the last time when the event trigger is met, e(t) denotes the error between the theoretical output torque of the motor calculated by the steering execution motor control law at the current time and the output torque of the motor at the last trigger time, and v1 and m1 are set constants, which denote the size of the error interval in the event trigger condition. When the error does not exceed the interval, the steering execution motor will always output the torque at the last time.

[0037] Further, the introduction of the Nussbaum function can eliminate the influence of the steering actuator failure on the control system, N(ξ) is the introduced Nussbaum function, and N(ξ) = ξ 2 cos(ξ), wherein ξ satisfies the following adaptive rate:

[0038]

[0039] wherein c2、 and ε are adjustable positive design parameters, which respectively represent the disturbance rejection ability of the sliding mode surface, the maximum error caused by the model, and the speed of the sliding mode surface tending to the limit, represents the actual estimation of the total disturbance torque by the neural network, satisfies the following adaptive rate:

[0040]

[0041] wherein γ and k are adjustable design parameters, which represent the Lyapunov parameters required to ensure system stability.

[0042] Further, the control rate can make the system state converge through the Lyapunov function V(t), and the formula of the Lyapunov function is:

[0043]

[0044] wherein s1 is a tracking error, representing the error between the steering wheel angle and the expected angle, s2 is a derivative of the tracking error, representing the error of the steering wheel angle speed, and v represents the output error of the filter, represents the error between the estimation and the actual value of the total disturbance torque by the neural network.

[0045] Further, in step S3, the control rate parameters include total rotation torque J eq , total viscous damping coefficient B eq , steering wheel kingpin rotation inertia J f , actuator rotation inertia J m2 , steering wheel kingpin damping coefficient B f , actuator damping coefficient B m2 , and transmission ratio K c between the motor angle and the steering wheel angle, and the adjustable parameter values in the steer-by-wire actuator control rate are adjusted according to the application environment, including the disturbance rejection ability c2 of the sliding mode surface, the maximum error caused by the model The size of the error interval v1 in the event triggering condition, m1, the Lyapunov parameter gamma required for ensuring system stability, and k are obtained, and the optimal values of these parameters are obtained, so that the Nussbaum function and the adaptive rate of the neural network are obtained, and the Nussbaum function N(ξ) is substituted into the Nussbaum function N(ξ) = ξ 2 cos(ξ) and the neural network calculation result Online real-time update their values, introduce them into the output torque formula of the motor, obtain the theoretical output torque p(t) of the motor, and finally determine whether the output torque at this moment meets the event triggering condition, if yes, update the output torque of the motor, if not, still maintain the original output torque of the motor.

[0046] The application also provides a storage medium of the event-triggered steer-by-wire angle tracking control method, which can be a ROM, a RAM, a magnetic disk, an optical disk or the like, and stores one or more programs, which can realize the event-triggered steer-by-wire angle tracking control method when executed by a processor.

[0047] Further, the application also discloses a device of the event-triggered steer-by-wire angle tracking control method, which can be a desktop computer, a notebook computer, a smart phone, a PDA handheld terminal, a tablet computer or other terminal devices with display function, and comprises a processor and a memory, the memory stores one or more programs, and the processor can realize the event-triggered steer-by-wire angle tracking control method when executing the program stored in the memory.

[0048] Compared with the prior art, the application has the following beneficial effects:

[0049] 1. The Nussbaum function is introduced to process the steering execution motor fault phenomenon that may occur in the steer-by-wire angle tracking control, the neural network adopted has the ability to fit nonlinear torque interference, can cope with more complex and severe working conditions, and has certain fault tolerance capability.

[0050] 2. The steering execution motor based on the event triggering mechanism is designed, the output of the steering execution motor is set with appropriate event triggering conditions, the occupancy rate of communication resources is reduced, the basic performance of the steer-by-wire angle tracking control is ensured, the dynamic response speed is fast, and the tracking precision is high. BRIEF DESCRIPTION OF DRAWINGS

[0051] Figure 1 The flowchart of the event-triggered steer-by-wire angle tracking control method is shown in the figure.

[0052] Figure 2 The reference diagram of the steer-by-wire system execution dynamics model is shown in the figure.

[0053] Figure 3A control structure diagram of the steer-by-wire system control method of the present application;

[0054] Figure 4 An effect diagram of the angle tracking of the steer-by-wire system under the condition of failure of the steering execution motor;

[0055] Figure 5 A tracking error broken line diagram of the angle tracking of the steer-by-wire system under the condition of failure of the steering execution motor;

[0056] Figure 6 An output torque broken line diagram of the steering motor of the steer-by-wire system under the condition of failure of the steering execution motor;

[0057] Figure 7 An event trigger interval diagram of the output of the steering motor of the steer-by-wire system under the condition of failure of the steering execution motor.

[0058] In the drawings:

[0059] 1 - steering wheel; 2 - steer-by-wire mechanism; 3 - steering wheel. DETAILED DESCRIPTION

[0060] The present application will be further described below in conjunction with specific embodiments. In the drawings, only exemplary descriptions are used, and it should not be understood as a limitation of the present patent; in order to better illustrate the embodiments, some components in the drawings may be omitted, enlarged or reduced, and do not represent the actual size of the product; for those skilled in the art, it is understandable that some well-known structures and their descriptions in the drawings may be omitted.

[0061] The same or similar reference numerals in the drawings of the embodiments of the present application correspond to the same or similar components; in the description of the present application, it should be understood that if the terms "front", "back", "left", "right" and the like indicate the orientation or positional relationship shown in the drawings, they are only used for the convenience of describing the present application and simplifying the description, and do not indicate or imply that the devices or elements referred to must have a particular orientation, be constructed and operated in a particular orientation, therefore the terms describing the positional relationship in the drawings are only used for exemplary description, and cannot be understood as a limitation of the present patent, for those skilled in the art, the specific meaning of the above terms can be understood according to the specific circumstances. In addition, in the present application, the description of "first", "second" and the like is only for the purpose of description, and cannot be understood as indicating or implying the relative importance of the technical features indicated or the number of technical features indicated. Therefore, the features limited by "first", "second" can explicitly or implicitly include at least one of the features.

[0062] Embodiment one:

[0063] Reference is made to Figure 1 , Figure 2 andFigure 3 The embodiment provides a kind of based on event trigger's steer-by-wire corner tracking control method, application is on steer-by-wire 2, the steer-by-wire 2 is using steering sensor monitoring steering wheel 1 and the rotation angle of car steering wheel 3, to cancel the mechanism of entity rotation shaft.The control method described in the embodiment first introduces the nonlinear variable in the simulation model of radial basis function neural network algorithm, determines the input, state and output of control system, establishes the dynamics model of steer-by-wire system execution mechanism.Then, the dynamics model established is designed for the corner tracking controller, introduces dynamic sliding surface method, Nussbaum function and event trigger condition, sliding mode control rate is derived using backstepping method, so that system converges in finite time, meet Lyapunov stability criterion.When the failure type fault or bias type fault of execution motor occurs, steer-by-wire 2 can monitor the rotation angle error of steering wheel 1 and car steering wheel 3.

[0064] When the corner error is greater than the set value, the steering execution motor can power compensation to steering wheel 3, so that the rotation angle of steering wheel 3 corresponds to the rotation angle of steering wheel 1 controlled by the driver, and the steering precision effect is achieved.At the same time, in order to save communication resources and avoid occupying too much communication resources and reducing the corner tracking precision, the embodiment also designs event trigger condition, so that the time of steering execution motor output produces interval, only the output torque of steering execution motor meets the set condition, otherwise the torque output will be always according to the torque output at last time.This can effectively reduce the torque output frequency of steering execution motor, and the output torque of steering execution motor under event trigger mechanism is discontinuous, so as to achieve the purpose of saving the occupancy rate of communication resources.

[0065] Then, specific control rate parameters are selected, the weight of neural network in control rate and Nussbaum function are updated, the torque control rate of steer-by-wire system execution mechanism is obtained by combining the sliding mode control rate, and the steering motor torque is calculated.Finally, steering wheel is turned according to the expected path by controlling steering execution motor torque.

[0066] The embodiment also includes a storage medium capable of realizing the above-mentioned event trigger-based steer-by-wire corner tracking control method, and the storage medium stores a program, which can realize the above-mentioned control method when executed by a processor.

[0067] Embodiment two:

[0068] On the basis of embodiment one, in the embodiment, the dynamics model of steer-by-wire system 2 execution mechanism can be expressed by the following formula:

[0069]

[0070] In the formula, Jeq denotes the total rotational moment, which can be equivalent to where J f denotes the kingpin rotational inertia of the steered wheel, K c denotes the transmission ratio between the motor angle and the steered wheel 3 angle, J m2 denotes the motor rotational inertia;

[0071] B eq denotes the total viscous damping coefficient, which can be equivalent to where B f denotes the kingpin damping coefficient of the steered wheel, K c denotes the transmission ratio between the motor angle and the steered wheel 3 angle, B m2 denotes the motor damping coefficient;

[0072] denotes the total moment disturbance, θ f , denote the actual rotational angle and the angular velocity of the steered wheel 3, respectively, which can be expressed as τ e +τ f , τ e is the tire self-aligning moment, τ f is the equivalent friction moment on the kingpin; T m denotes the output of the motor.

[0073] In order to better handle the moment disturbance in the above dynamic model, the present embodiment adopts a radial basis neural network to approximate that is, it is assumed to be equal to where, denotes the input node of the radial basis neural network, denotes the output weight of each hidden layer, the superscript T represents transposition, and the vector φ(x) ∈ R N is a Gaussian type function, which is defined as:

[0074]

[0075] In the formula, g i and σ i are the center value and width of the Nth neural network node, respectively.

[0076] Then, the steering actuator torque is defined as the system input, the steered wheel angle and speed are defined as the system state, and the actual steered wheel angle is defined as the system output, while considering the actuator fault in the system output, to obtain the state space expression of the steering actuator:

[0077]

[0078] In the formula, define and G(t) = β(t-T)g(x ), g( x ) is a fault burst at time β(t-T), and T n is the time when the fault occurs, when t n , G(t) = 1, when t ≥ T n , G(t) = 1 + g( x ), u is the control input value ε1 represents the error of the neural network estimation, and d(t) represents unknown external disturbance; in the formula, x1 represents the actual rotation angle of the steering wheel, x2 represents the angular velocity of the steering wheel, and y represents the output of the motor. The control target of the steering angle tracking controller of the steer-by-wire system 2 is to eliminate the error between the actual steering angle of the controlled vehicle and the expected steering angle, so as to realize steering angle tracking control.

[0079] Embodiment three:

[0080] On the basis of embodiment two, the backstepping method is selected as the main control method in this embodiment. In order to solve the inherent term expansion problem in the backstepping method, a dynamic sliding mode surface method is also introduced in this embodiment, and an integral filter is used to calculate the virtual control reciprocal, so as to eliminate the expanded terms, and the design of the filter is as follows:

[0081]

[0082] In the formula, τ is a time constant, represents the time derivative of the output, represents the output of the first-order low-pass filter a, which is calculated by the following formula:

[0083]

[0084] In the formula, s1 is the tracking error, which represents the error between the steering wheel angle and the expected steering wheel angle, s2 is the derivative of the tracking error, which represents the error of the steering wheel angular velocity, and v represents the output error of the filter. The above error values can be monitored by using a steering sensor.

[0085] On the basis of the above, the control law of the steering execution motor is designed by using the filter:

[0086]

[0087] In the formula, p(t) represents the torque output value of the steering execution motor, N(ξ) is the Nussbaum function, c2、 and ε are adjustable positive value parameters designed, which respectively represent the disturbance rejection ability of the sliding mode surface, the maximum error of the model, and the speed of the sliding mode surface tending to the limit, represents the actual estimation value of the total disturbance torque by the neural network.

[0088] In order to reduce the occupation rate of communication resources, the torque output value is set with an event trigger condition, so that the output torque of the motor is updated only when the condition is met, otherwise the torque output of the last time is always followed:

[0089]

[0090] In the formula, u(t) represents the actual output torque of the steering execution motor, p(t k ) represents the output torque of the steering motor at the last time when the event trigger is met, e(t) represents the error between the theoretical output torque of the motor calculated by the steering execution motor control rate at the current time and the motor output torque at the last trigger time, v1 and m1 are set constants, and represent the size of the error interval in the event trigger condition. When the error does not exceed the interval, the steering execution motor always follows the torque output at the last time.

[0091] At the same time, in order to eliminate the influence of the steering execution motor failure on the control system, the Nussbaum function N(ξ) = ξ 2 cos(ξ) is introduced, in which ξ satisfies the following adaptive rate:

[0092]

[0093] In the formula, c2、 and ε are adjustable positive value parameters designed, which respectively represent the disturbance rejection ability of the sliding mode surface, the maximum value of the error brought by the model, and the speed of the sliding mode surface tending to the limit, represents the actual estimation of the total disturbance torque by the neural network, satisfies the following adaptive rate:

[0094]

[0095] In the formula, γ and k are adjustable parameters designed, which represent the Lyapunov parameter required to ensure system stability. Through the Lyapunov function it can be judged that the control rate can make the system state converge. The backstepping dynamic sliding mode control is a control algorithm with strong robustness and strong anti-interference ability, the neural network used has the ability to fit nonlinear torque disturbance, the Nussbaum function is introduced to process the failure of the execution motor, and the designed event trigger condition can fully save the communication resources.

[0096] Example four:

[0097] According to the event-triggered steer-by-wire corner tracking control method disclosed in Embodiment Two and Embodiment Three, the present embodiment further discloses a device capable of implementing the above control method, which comprises a processor and a memory, the memory stores a plurality of programs, and the processor can implement the above control method when executing the programs stored in the memory. The main technical performance indicators and device parameters used by the device include:

[0098] The disturbance rejection capability c2 of the sliding surface = 10, the maximum value of the error brought by the model The size of the error interval in the event-triggering condition v1 = 0.16 and m1 = 0.2, the Lyapunov parameter γ = 1000 and k = 0.1 required to ensure system stability, the steering wheel kingpin moment of inertia J f = 1.635 (kg·m 2 ), the actuator motor moment of inertia J m2 = 0.0451 (kg·m 2 ), the damping coefficient of the steering wheel kingpin B f = 10 (N·m / (rad / s)), the damping coefficient of the actuator motor B m2 = 0.05 (N·m / (rad / s)), and the transmission ratio K c between the motor angle and the steering wheel angle = 30.

[0099] Substituting the above parameters into the formulas of Embodiment Two and Embodiment Three, the actual response of the steering actuator motor under the condition of a 10% failure type fault superimposed with a 10 N·M bias type fault is obtained, as shown in Figure 4 The actual steering angle (deg) of the steering wheel 3 and the reference angle (i.e. the expected angle) exist an error less than 0.1 seconds, that is, when the driver turns the steering wheel 1, the steer-by-wire mechanism 2 will transmit the turning signal to the steering wheel 3 within 0.1 seconds.

[0100] Referring to Figure 5 and Figure 6 , Figure 5 the error of the corner tracking is given, Figure 6 the actual output torque of the steering actuator motor in these examples is given. Figure 5 The curve in the figure represents the corner tracking error, when the error is greater than the set limit, the output torque of the steering actuator motor will be dynamically adjusted to control the corner tracking error. At the same time, in order to save communication resources, the steering actuator motor is provided with an event-triggering condition, so that the output torque of the motor will only be updated when the condition is met, otherwise it will always output the torque at the last moment, the output torque of the steering actuator motor is not continuous, which is reflected in Figure 6The turning angle tracking error reaches the maximum value in the current period when the peak or valley value of the turning angle tracking error appears with the increase of the output torque variation of the motor.

[0101] Figure 7 The time interval of the steering execution motor output under the event trigger condition is given. The trigger interval of the traditional time trigger condition is fixed (usually 0.02 seconds, that is, 50 Hz), so it will be triggered 50 times in 1 second, which means 50 times of data transmission. The trigger interval of the event trigger condition is dynamic, and the maximum interval can reach 0.11 seconds. Compared with the time trigger condition, the trigger interval of the event trigger condition is relatively longer, so the number of triggers in the same time is reduced, and the output torque of the steering execution motor is not continuous. Therefore, the occupation rate of communication resources can be fully saved, the turning angle tracking error is small, and the tracking precision is high.

[0102] In the specific content of the above specific embodiments, each technical feature can be combined arbitrarily without contradiction. To make the description concise, all possible combinations of the above technical features are not described, but as long as the combination of the technical features does not exist contradiction, it should be considered as the scope of the present application.

[0103] Obviously, the above embodiments of the present application are only examples for clearly illustrating the present application, and are not intended to limit the embodiments of the present application. Based on the above description, other different forms of changes or variations can be made by those skilled in the art. Here, it is not necessary and impossible to exhaust all the embodiments. Any modification, equivalent replacement and improvement made within the spirit and principle of the present application should be included in the protection scope of the claims of the present application.

Claims

1. A method for drive-by-wire corner tracking control based on event triggering, characterized in that, The method includes the following steps: Step S1: Introduce nonlinear variables in the radial basis function neural network algorithm simulation model, determine the input, state and output of the control system, and establish the dynamic model of the actuator of the steer-by-wire system; Step S2: Design an angle tracking controller for the dynamic model established in Step S1. Introduce the dynamic sliding surface method, Nussbaum type function, and event triggering conditions. Use the backstepping method to derive the sliding control law, making the system converge in finite time and satisfying the Lyapunov stability criterion. The dynamic sliding surface method described in step S2 includes the following steps: Step S21: Calculate the reciprocal of the virtual control using an integral filter to eliminate the inflation term. The filter is designed as follows: In the formula, It is a time constant, which determines the filter's response speed; α represents the virtual control input signal. This represents the value of the input signal α after low-pass filtering, and represents a smoothed estimate of the original signal. It is the output signal The derivative with respect to time represents the rate of change of the output, and α(0) represents the initial value of the filter input signal α at the initial time. Indicates the filter output signal The initial value at the initial moment, α and It is calculated using the following formula: In the formula, e and s1 are tracking errors, representing the error between the steering wheel angle and the desired steering angle; s2 is the derivative of the tracking error, representing the steering wheel angular velocity error; x2 represents the steering wheel angular velocity; and v represents the filter output error. All of the above error values ​​can be obtained by monitoring the steering sensor. Step S22: Design the control law for the steering actuator motor using the filter: In the formula, ρ(t) represents the torque output value of the steering actuator motor, N(ξ) is the Nussbaum function, and c2, And ε is an adjustable positive parameter of the design, which respectively represents the disturbance rejection capability of the sliding surface, the maximum error introduced by the model, and the speed at which the sliding surface approaches its limit. This represents the actual estimated value of the total disturbance torque by the neural network; Step S23: The following event triggering condition is set for the torque output value, so that the motor output torque will only be updated when the condition is met; otherwise, it will continue to output the torque according to the previous moment: In the formula, u(t) represents the actual output torque of the steering actuator motor, and ρ(t) k ) represents the output torque of the steering motor at the previous event triggering moment, e(t) represents the error between the theoretical output torque of the motor calculated by the steering motor control rate at the current moment and the output torque of the motor at the previous triggering moment, v1 and m1 are set constants, representing the size of the error range in the event triggering condition. When the error does not exceed its range, the steering motor will continue to output torque according to the previous moment. Step S3: Select specific control rate parameters, update the weights of the neural network and the Nussbaum function in the control rate of step S2, and combine the sliding mode control rate to obtain the torque control rate of the actuator of the steer-by-wire system, and calculate the torque of the steering motor. Step S4: Control the torque of the steering actuator motor to make the steering wheel turn along the expected path.

2. The event-triggered steerable corner tracking control method according to claim 1, characterized in that, The dynamic model of the actuator of the steer-by-wire system described in step S1 can be expressed by the following formula: In the formula, J eq This represents the total torque, which can be equivalent to... J f K represents the moment of inertia of the kingpin of the steering wheel. c J represents the transmission ratio between the motor rotation angle and the steering wheel rotation angle. m2 Indicates the moment of inertia of the motor. B eq This represents the total viscous damping coefficient, which can be equivalent to... Among them B f K represents the damping coefficient of the steering wheel kingpin. c B represents the transmission ratio between the motor rotation angle and the steering wheel rotation angle. m2 Indicates the damping coefficient of the motor; θ represents the total torque disturbance. f , These represent the actual rotation angle and angular velocity of the steering wheel, respectively, and can be expressed as τ. e +τ f , τ e τ is the self-aligning torque of the tire. f Equivalent frictional torque on the main pin; T m This indicates the output of the motor.

3. The event-triggered steerable corner tracking control method according to claim 2, characterized in that, In step S1, a radial basis function neural network is introduced to calculate the actual value of the total torque using the following formula: In the formula, Let w(X) represent the input node of the radial basis function neural network, Φ represent the output weights of each hidden layer, the superscript T represents the transpose, and the vector w(X) ∈ R. N It is a Gaussian function, defined as: In the formula, g i and σ i These are the center value and width of the i-th neural network node, respectively.

4. The event-triggered steerable corner tracking control method according to any one of claims 1-3, characterized in that, In step S1, the torque of the steering actuator motor is defined as the system input, the steering wheel angle and speed are defined as the system state, and the actual steering wheel angle is defined as the system output. Simultaneously, actuator faults are considered in the system output, resulting in the state-space expression for the steering actuator: In the formula, it is defined And G(t)=β(tT)g( x ), β(tT) represents the fault function, where t is time and T is the moment the fault occurs, g( x The fault is a sudden fault that occurs at time β(tT), where T is defined as... n When the fault occurs, t <T n When t ≥ T, G(t) = 1. n At that time, G(t) = 1 + g( x u is the control input value. Φ T w(X) is the output of the radial neural network, ε1 represents the estimation error of the neural network, and d(t) represents the unknown external disturbance. x1 represents the actual rotation angle of the steering wheel, x2 represents the angular velocity of the steering wheel, and y represents the output of the motor.

5. The event-triggered steerable corner tracking control method according to claim 1, characterized in that, Introducing the Nussbaum function can eliminate the impact of a steering actuator motor failure on the control system. N(ξ) is the introduced Nussbaum function, and N(ξ) = ξ 2 cos(ξ), where ξ satisfies the following adaptive law: In the formula, c2, And ε is an adjustable positive parameter of the design, which respectively represents the disturbance rejection capability of the sliding surface, the maximum error introduced by the model, and the speed at which the sliding surface approaches its limit. This represents the neural network's estimate of the total disturbance torque. Satisfying the following adaptive rate: In the formula, γ and k are the adjustable parameters of the design, which represent the Lyapunov parameters required to ensure the stability of the system, and w(X) is the weight function of the radial neural network.

6. The event-triggered steerable corner tracking control method according to claim 5, characterized in that, The Lyapunov function V(t) can be used to determine whether the control law can bring the system state to convergence. The formula for the Lyapunov function is as follows: In the formula, s1 is the tracking error, representing the error between the steering wheel angle and the desired steering angle; s2 is the derivative of the tracking error, representing the steering wheel angular velocity error; and v represents the filter output error. This represents the error between the neural network's predicted and actual total disturbance torque values, where γ is an adjustable parameter in the design.

7. The event-triggered steerable corner tracking control method according to claim 1, characterized in that, In step S3, the control rate parameter includes the total torque J. eq Total viscous damping coefficient B eq Steering wheel kingpin moment of inertia J f , Moment of inertia of the motor J m2 Damping coefficient B of the steering wheel kingpin f , Motor damping coefficient B m2 And the transmission ratio K between the motor rotation angle and the steering wheel rotation angle c The adjustable parameter values ​​in the control rate of the steering-by-wire actuator are adjusted according to the application environment, including the anti-disturbance capability c2 of the sliding surface and the maximum value of the error introduced by the model. The error intervals v1 and m1 in the event triggering conditions, the Lyapunov parameters γ and k required to ensure system stability, and the optimal values ​​of these parameters will yield the Nussbaum function and the adaptive rate of the neural network. Substituting these values ​​into the Nussbaum function N(ξ)=ξ 2 cos(ξ) and neural network calculation results The values ​​are updated online in real time and introduced into the motor output torque formula to obtain the theoretical output torque ρ(t) of the motor. Finally, it is determined whether the output torque at that moment meets the event triggering condition. If it does, the motor output torque is updated; otherwise, the original steering motor output torque is maintained.

8. A storage medium for an event-triggered drive-by-wire cornering tracking control method, which stores one or more programs, characterized in that, When the program is executed by the processor, it can implement the event-triggered steerable cornering tracking control method as described in any one of claims 1-7.

9. A device for an event-triggered drive-by-wire cornering tracking control method, comprising a processor and a memory, wherein the memory stores one or more programs, characterized in that, When the processor executes the program stored in the memory, it can implement the event-triggered steerable cornering tracking control method as described in any one of claims 1-7.

Images