Model-free event differential control method and device, storage medium and equipment

Abstract

The invention provides a model-free event differential control method and device, a storage medium and equipment. The model-free event differential control method comprises the steps that an expected signal, a real-time input signal and a real-time output signal of a vehicle are obtained; constructing an input and output dynamic model of the motor system according to the historical output signal and the historical input signal of the vehicle motor system; obtaining a fully-formatted dynamic linearization model of the vehicle motor system according to the input and output dynamic model; obtaining an estimated pseudo partial derivative according to a historical input signal, a historical output signal and the pseudo partial derivative initial value of the vehicle motor system; determining the following error according to the expected signal and the real-time output signal, and determining a motor input signal difference value according to the following error and the pseudo gradient in combination with an RBF neural network controller; and if the vehicle meets the tracking error threshold condition or the event triggering condition, updating the input signal of the vehicle motor according to the motor input signal difference value.

Description

Technical Field

[0001] The present invention relates to the field of vehicle control, and particularly to a model-free event differential control method, device, storage medium and equipment. Background Art

[0002] With the rapid popularization of new energy vehicles, the importance of electric drive control technology for vehicle performance has gradually emerged. The electric drive control system can accurately control the speed of the motor to ensure that the vehicle can obtain the best power output under different working conditions. By optimizing the electric drive control strategy, the vibration and noise of the vehicle can be reduced, and the stability of the vehicle can be improved. Electric drive control plays an important role in improving vehicle performance, enhancing vehicle safety, and improving vehicle stability.

[0003] The Ackermann steering structure adjusts the steering angle difference between the inner wheel and the outer wheel, so that the steering centers of all wheels intersect at a point on the extension line of the rear axle (i.e., the turning center), thereby realizing an approximate pure rolling steering. This structure can optimize the tire movement trajectory during vehicle turning, reduce tire side slip or wear, and improve steering stability and tire life.

[0004] Ackermann transport equipment usually uses the method of driving the rear wheels by motors to realize the forward, backward and differential control of the vehicle. The front wheels control the steering angle through a servo motor to realize the steering function of the vehicle. For the motor control method of Ackermann vehicles, PID control, fuzzy control and neural network control methods are mostly used. The steering angle signal and the speed signal are input into the Ackermann steering model to obtain the rotational speeds of the left rear wheel and the right rear wheel and the rotation angle of the servo motor. Then, the actual wheel rotational speed and the servo motor steering angle are measured by sensors, and the difference between the two is used as a control variable. Through relevant control algorithms, the drive motor current is adjusted to make the actual vehicle speed follow the ideal value to realize the coordinated rotational speed control of each drive motor.

[0005] With the rapid development of unmanned driving technology, a large number of sensors are installed on the vehicle body to provide data support for increasingly complex unmanned driving algorithms; however, the existing motor drive methods only focus on the rotational speed and following accuracy, ignoring the impact of the algorithm occupying a large amount of computing power of the device. In this regard, there is an urgent need for a motor control method that takes into account both control performance and reduces the update frequency of the controller to reduce the computing amount of the device, so as to provide support for the implementation of Ackermann transport equipment in real-world scenarios. Summary of the Invention

[0006] Based on this, the present invention provides a model-free event differential control method, device, storage medium and equipment, which are directly applied to the unknown motor system of Ackermann vehicles, and only use the input / output data during the operation of the vehicle motor system to adjust the input control signal of the vehicle at the next moment, and combine the event trigger condition to perform differential control on the vehicle to achieve the target motion trajectory of the vehicle.

[0007] In a first aspect, the present invention provides a model-free event differential control method, including:

[0008] Obtain the desired signal, real-time input signal, and real-time output signal of the vehicle;

[0009] Construct an input-output dynamic model of the vehicle motor system based on the historical output signal and historical input signal of the vehicle motor system;

[0010] Obtain a fully formatted dynamic linearization model of the vehicle motor system according to the input-output dynamic model;

[0011] Obtain an estimated pseudo partial derivative according to the historical input signal, historical output signal, and initial value of the pseudo partial derivative of the vehicle motor system;

[0012] Determine the following error according to the desired signal and real-time output signal, and determine the difference of the motor input signal according to the following error and pseudo gradient in combination with the RBF neural network controller;

[0013] If the vehicle meets the tracking error threshold condition or event trigger condition, update the input signal of the vehicle motor according to the difference of the motor input signal.

[0014] Further, the specific expression for constructing the input-output dynamic model of the vehicle motor system based on the historical output signal and historical input signal of the vehicle motor system is:

[0015] ,

[0016] Among them, the specific expression of the historical output signal of the vehicle motor system is , is the actual rotational speed of the left rear wheel of the vehicle at time is the actual rotational speed of the right rear wheel of the vehicle at time is the actual steering angle of the vehicle's steering servo at time; the specific expression of the historical input signal of the vehicle motor system is , is the left rear wheel PWM control signal of the vehicle motor system at time is the right rear wheel PWM control signal of the vehicle motor system at time is the steering servo PWM control signal of the vehicle motor system at time; is an unknown continuously differentiable function, is the first positive definite system order, is the second positive definite system order.

[0017] Furthermore, the specific expression of the full-format dynamic linearization model is as follows:

[0018] ,

[0019] wherein, is the change amount of the input and output of the vehicle motor system at time , is the pseudo partial derivative of the vehicle motor system at time .

[0020] Furthermore, the obtaining of the predicted pseudo partial derivative according to the historical input signal, historical output signal and initial value of the pseudo partial derivative of the vehicle motor system includes:

[0021] Step S201, let , set the preset data length and the initial value of the pseudo partial derivative ;

[0022] Step S202, obtain the historical input signal of the vehicle motor system and the historical output signal at the corresponding time;

[0023] Step S203, input the initial value of the pseudo partial derivative, the historical input signal of the vehicle motor system and the historical output signal at the corresponding time into the specific expression of the predicted pseudo partial derivative, and obtain the predicted pseudo partial derivative ;

[0024] Step S204, calculate the output prediction value of the Ackermann vehicle motor system at the next moment based on the predicted pseudo partial derivative, and obtain the prediction error according to the output prediction value and the actual output value at the next moment. If the absolute value of the prediction error is less than the preset threshold, output the predicted pseudo partial derivative; otherwise, update to , and repeat steps S202 - S204 until the absolute value of the prediction error is less than the preset threshold.

[0025] Furthermore, the specific expression of the motor input signal difference is as follows:

[0026] ,

[0027] wherein, is the motor input signal difference, is the set of connection weights between each hidden neuron and the output layer, , is the set of output values of the RBF neural network hidden neurons, , is the number of output layers.

[0028] Furthermore, when the vehicle meets the tracking error threshold condition or the event trigger condition, updating the input signal of the vehicle motor according to the motor input signal difference is specifically as follows:

[0029] When and meets ,

[0030] When and meets or ,

[0031] Update the input signal of the vehicle motor to:

[0032] ;

[0033] where the event trigger condition is or , is the event trigger error, , is the predicted output signal, , is the first parameter, , is the second parameter, , is the estimated pseudo partial derivative at time is the actual output signal, .

[0034] In a second aspect, the present invention also provides a model-free event differential control device, including:

[0035] A signal acquisition module for acquiring the desired signal, real-time input signal, and real-time output signal of the vehicle;

[0036] An input-output model construction module for constructing an input-output dynamic model of the motor system according to the historical output signal and historical input signal of the vehicle motor system;

[0037] A model conversion module for obtaining a fully formatted dynamic linearization model of the vehicle motor system according to the input-output dynamic model;

[0038] A pseudo partial derivative estimation module for obtaining an estimated pseudo partial derivative according to the historical input signal, historical output signal, and initial value of the pseudo partial derivative of the vehicle motor system;

[0039] An input signal difference determination module, configured to determine a following error according to the desired signal and the real-time output signal, and determine the difference of the motor input signal according to the following error and the pseudo-gradient in combination with an RBF neural network controller;

[0040] An input signal update module, configured to update the input signal of the vehicle motor according to the difference of the motor input signal if the vehicle meets the tracking error threshold condition or the event trigger condition.

[0041] Further, the pseudo partial derivative estimation module includes:

[0042] A parameter setting unit, configured to make , set a preset data length and the initial value of the pseudo partial derivative ;

[0043] A historical signal acquisition unit, configured to acquire the historical input signal of the vehicle motor system and the historical output signal at the corresponding moment;

[0044] A pseudo partial derivative estimation unit, configured to input the initial value of the pseudo partial derivative, the historical input signal of the vehicle motor system, and the historical output signal at the corresponding moment into the specific expression for estimating the pseudo partial derivative to obtain the estimated pseudo partial derivative ;

[0045] A predicted pseudo partial derivative correction output unit, configured to calculate the predicted output value of the Ackermann vehicle motor system at the next moment based on the predicted pseudo partial derivative, and obtain a prediction error according to the predicted output value and the actual output value at the next moment , if the absolute value of the prediction error is less than a preset threshold, output the predicted pseudo partial derivative; otherwise, make updated to , and repeat steps S202 - S204 until the absolute value of the prediction error is less than the preset threshold.

[0046] In a third aspect, the present invention further provides a computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, the steps of the model-free event differential control method in any one of the first aspects are implemented.

[0047] In a fourth aspect, the present invention further provides a computer device, including a memory and a processor, where the memory stores a computer program, and when the processor executes the computer program, it executes the model-free event differential control method in any one of the first aspects.

[0048] The beneficial effects of adopting the above technical solutions are as follows: The model-free adaptive event-triggered control proposed in this embodiment is directly applied to the unknown motor system of an Ackermann vehicle. The event-triggering condition is calculated only using the input / output (I / O) data during the operation of the motor system. Therefore, the proposed method is data-based and does not rely on any explicit motor system model information.

[0049] Moreover, the control method proposed in this embodiment can reduce the execution times of the controller and save computational costs while ensuring system stability and control performance.

[0050] In addition, the offline parameter identification algorithm is integrated into the event-triggered model-free adaptive control method, and the obtained data model is continuously updated online using input / output data, so as to adapt to changes in system dynamics and disturbances. This improves vehicle stability. BRIEF DESCRIPTION OF THE DRAWINGS

[0051] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for the description of the embodiments or the prior art.

[0052] Figure 1 Schematic diagram of the model-free event differential control method in an embodiment of this application;

[0053] Figure 2 Schematic diagram of the kinematic model of an Ackermann vehicle in an embodiment of this application;

[0054] Figure 3 Schematic diagram of the flow of the model-free event differential control method in an embodiment of this application;

[0055] Figure 4 Schematic diagram of the model-free event differential control device in an embodiment of this application. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0056] The following will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention. In order to describe the present invention in more detail, the model-free event differential control method, device, storage medium, and equipment provided by the present invention will be specifically described below with reference to the drawings.

[0057] Unless otherwise defined, the technical terms or scientific terms used in this application disclosure shall have the ordinary meanings understood by those of ordinary skill in the art to which this invention pertains. The "first", "second" and similar terms used in this invention do not denote any order, quantity or importance, but are only used to distinguish different components. Similarly, words such as "a", "an" or "the" do not denote a quantity limitation, but mean that there is at least one. Words such as "comprising" or "including" mean that the elements or objects appearing before this word cover the elements or objects listed after this word and their equivalents, without excluding other elements or objects. Words such as "connected" or "coupled" are not limited to physical or mechanical connections, but may include electrical connections, whether direct or indirect. "Upper", "lower", "left", "right", etc. are only used to indicate relative position relationships. When the absolute position of the object being described changes, the relative position relationship may also change accordingly.

[0058] In the existing centralized drive mode, the structure and function of the spur cylindrical gear differential are similar to those of the ordinary open differential applied in traditional fuel vehicles. When the vehicle enters a split road surface, due to the characteristic of torque equal distribution of the spur cylindrical gear differential, the vehicle will experience slipping and instability, or even be unable to continue driving, reducing the passing performance of the vehicle. Further, in the existing centralized drive electric vehicles, the differential performance of the inner and outer wheels depends on the design of the mechanical differential structure. In distributed drive electric vehicles, each drive motor is independently distributed around the vehicle, and the two-wheel drive eliminates the traditional vehicle differential, simplifying the vehicle structure.

[0059] For the traditional control methods of the electronic control system, the PID control method is difficult to meet the requirements of high-efficiency and high-precision control of the in-wheel motor due to the lack of the ability to handle system time-varying disturbances. Fuzzy control may result in relatively low control precision of the system due to its fuzzy processing mechanism, and the steady-state precision is generally not high. The design of a fuzzy controller often depends on the experience and knowledge of the designer, including the formulation of fuzzy rules and the selection of membership functions. These design factors largely determine the performance of the controller, but also increase the complexity and subjectivity of the design. Once the fuzzy control rules are determined, its flexibility is relatively poor, and it is not easy to modify and adjust online. The existing motor control methods of the electronic control system waste computing power resources because, due to the lack of an effective event detection and response mechanism, the system has to rely on a fixed time period to operate and continuously performs calculations even when there is no significant state change.

[0060] In response to this, the present invention provides a model-free event differential control method, device, storage medium, and equipment, which are applied to an Ackermann vehicle with an unknown motor model and only rely on the data measured in real time by the controlled system without relying on any mathematical model of the controlled system. This method does not require any external test signals or training processes, has the advantages of small computational burden, easy implementation, and strong robustness, and has better performance compared with PID control, fuzzy control, or neural network control. Taking the application of this method to a terminal device as an example for illustration, please refer to the attached Figure 1 Schematic diagram of the model-free event differential control method shown.

[0061] The embodiments of the present application provide an application scenario of the model-free event differential control method. This application scenario includes the terminal device provided in the embodiment. The terminal device includes a right rear wheel speed controller, a left rear wheel speed controller, and a front wheel steering angle controller on an Ackermann vehicle. The form of the terminal device includes, but is not limited to, a smart phone and a computer device. Among them, the computer device can be at least one of devices such as a desktop computer, a portable computer, a laptop computer, a mainframe computer, and a tablet computer. The terminal device receives the current actual motion trajectory and the target motion trajectory of the vehicle and adjusts the motor input signal of the vehicle. For the specific process, please refer to the embodiments of the model-free event differential control method.

[0062] The following specifically describes the steps of the embodiment of the model-free event differential control method:

[0063] Step S101: Obtain the desired signal, real-time input signal, and real-time output signal of the vehicle.

[0064] Among them, in combination with the Figure 2 shown kinematic model of the Ackermann vehicle, the Ackermann vehicle usually uses the motor to drive the rear wheels to realize the forward and backward movement of the vehicle; in addition, the front wheels control the steering angle through the steering gear to realize the steering function of the vehicle. For the three-axis target speed of the Ackermann vehicle, it is denoted as , is the target speed of the vehicle on the X axis, is the target speed of the vehicle on the Y axis, is the target speed of the vehicle on the Z axis; the turning radius of the Ackermann vehicle is , the wheelbase is , and the wheelbase is .

[0065] Based on the kinematic model of the Ackermann vehicle, the specific drive wheel target speed and the front wheel target deflection angle of the Ackermann vehicle can be obtained according to the three-axis target speed of the vehicle and the structural parameters of the Ackermann vehicle (including the turning radius, wheelbase, and wheelbase); among them, the drive wheels of the Ackermann vehicle are two rear wheels, and the target speed of the left rear wheel is denoted as , the target speed of the left rear wheel is denoted as , the front wheels of an Ackermann vehicle are represented by the steering angle. The target steering angle of the left front wheel is denoted as , and the target steering angle of the right front wheel is denoted as .

[0066] The turning radius of the vehicle has the following specific expression:

[0067] ;

[0068] The specific expression for the target speed of the left rear wheel of the vehicle is:

[0069] ;

[0070] The specific expression for the target speed of the right rear wheel of the vehicle is:

[0071] ;

[0072] The specific expression for the target steering angle of the left front wheel of the vehicle is:

[0073] ;

[0074] The specific expression for the target steering angle of the right front wheel of the vehicle is:

[0075] .

[0076] The positive and negative values of the above target speed are defined according to the movement direction of the Ackermann vehicle. When the Ackermann vehicle is moving forward, the target speed is positive; when the Ackermann vehicle is moving backward, the target speed is negative. The positive and negative values of the above target steering angle are defined according to the movement direction of the Ackermann vehicle. When the Ackermann vehicle is turning left, the target steering angle is positive; when the Ackermann vehicle is turning right, the target steering angle is negative.

[0077] Furthermore, since this embodiment is based on an Ackermann steering structure with a crank-rocker mechanism, the steering wheel is driven by a driving servo, and the steering wheel then controls the steering of the two front wheels through mechanical transmission. Therefore, the steering angle of the steering wheel can be represented by the target steering angle of the front wheels of the vehicle, and the specific expression is:

[0078] ,

[0079] is the steering angle of the steering wheel.

[0080] Based on the above parameters, the desired signal of the vehicle can be denoted as , where is the target speed of the left rear wheel of the Ackermann vehicle at time , is The target speed of the right rear wheel of the Ackermann vehicle at a moment, , is the target steering angle of the front wheels of the Ackermann vehicle at a moment, ; The real-time output signal of the vehicle can be denoted as , is the actual rotation speed of the left rear wheel of the Ackermann vehicle at a moment, is the actual rotation speed of the right rear wheel of the Ackermann vehicle at a moment, is the actual rotation angle of the steering servo of the Ackermann vehicle at a moment; The real-time input signal of the vehicle can be denoted as , is the left rear wheel PWM control signal of the vehicle motor system at a moment, is the right rear wheel PWM control signal of the vehicle motor system at a moment, is the steering servo PWM control signal of the vehicle motor system at a moment.

[0081] Step S102, construct an input-output dynamic model of the motor system according to the historical output signals and historical input signals of the vehicle motor system.

[0082] In this embodiment, the rear drive wheels of the Ackermann vehicle adopt DC motors, and the microprocessor controls the rotation speed of the drive wheel motors by outputting PWM signals and controls the steering angle of the front wheel servo at the same time. The unknown motor system in the Ackermann vehicle can be regarded as a non-linear non-affine input-output dynamic model, that is, an input-output dynamic model of the motor system can be constructed according to the historical output signals and historical input signals of the motor system in the Ackermann vehicle. The specific expression is:

[0083] ,

[0084] Among them, the specific expression of the historical output signal of the vehicle motor system is , is the actual rotation speed of the left rear wheel of the Ackermann vehicle at a moment, is the actual rotation speed of the right rear wheel of the Ackermann vehicle at a moment, is the actual rotation angle of the steering servo of the Ackermann vehicle at a moment; The specific expression of the historical input signal of the vehicle motor system is , is the left rear wheel PWM control signal of the vehicle motor system at a moment, is The PWM control signal of the right rear wheel of the vehicle motor system at a moment, is the PWM control signal of the steering servo of the vehicle motor system at a moment. is an unknown continuously differentiable function, is the order of the first positive definite system, is the order of the second positive definite system.

[0085] For the historical input signals and historical output signals of the above vehicle motor system, for more concise description, use to represent the th input signal of the vehicle motor system at a moment, use to represent the th output signal of the vehicle motor system at a moment, , .

[0086] Step S103, obtain the full-format dynamic linearization model of the vehicle motor system according to the input-output dynamic model.

[0087] Specifically, with the help of the model-free adaptive dynamic linearization method, convert the input-output dynamic model obtained in the above step S102 into a full-format dynamic linearization model. The specific expression of the full-format dynamic linearization model is:

[0088] ,

[0089] where is the change in the input and output of the vehicle motor system at a moment, , is the pseudo partial derivative of the vehicle motor system at a moment, .

[0090] Step S104, obtain the predicted pseudo partial derivative according to the historical input signal, historical output signal and initial value of the pseudo partial derivative of the vehicle motor system.

[0091] Specifically, the pseudo partial derivative at a specific moment in the above step S103 cannot be directly obtained in the actual system operation. Therefore, a motor dynamic system identification method is designed to calculate with the predicted pseudo partial derivative at a specific moment instead of the actual value of the pseudo partial derivative. The specific steps are as follows:

[0092] Use to represent the predicted value of the pseudo partial derivative , that is, the predicted pseudo partial derivative, represents the derivative of The predicted value of the motor output of the Ackermann vehicle at a certain moment. Based on the fully formatted dynamic linearization model, the specific expression of the prediction model for the output change of the Ackermann vehicle motor system is as follows: , where the prediction error of this prediction model is , and the specific expression for estimating the pseudo partial derivative is thus:

[0093] .

[0094] Furthermore, the identification process of the estimated pseudo partial derivative includes the following specific steps:

[0095] Step S201, let , set the preset data length and the initial value of the pseudo partial derivative ;

[0096] Step S202, obtain the historical input signal of the vehicle motor system and the historical output signal at the corresponding moment;

[0097] Step S203, input the initial value of the pseudo partial derivative, the historical input signal of the vehicle motor system, and the historical output signal at the corresponding moment into the specific expression of the estimated pseudo partial derivative , to obtain the estimated pseudo partial derivative ;

[0098] Step S204, calculate the predicted value of the output of the Ackermann vehicle motor system at the next moment based on the estimated pseudo partial derivative, and obtain the prediction error according to the predicted value and the actual value of the output at the next moment , if the absolute value of the prediction error is less than the preset threshold, that is , is the preset threshold, , then output the estimated pseudo partial derivative; otherwise, update to , and repeat steps S202 - S204 until the absolute value of the prediction error is less than the preset threshold.

[0099] Step S105, determine the following error according to the desired signal and the real - time output signal, and determine the difference of the motor input signal according to the following error and the pseudo - gradient combined with the RBF neural network controller.

[0100] Specifically, in this embodiment, the motor model of the Ackermann vehicle is an ideal non - linear controller, and the specific expression of the vehicle motor model is:

[0101] ,

[0102] where, is the difference operator, is The pseudo-gradient at a moment , is , the following error , which represents the error value between the actual trajectory and the target trajectory of the vehicle at a moment, , , and

[0103] Furthermore, use the RBF neural network to approximate . The output value of the th hidden neuron of the RBF neural network is:

[0104] ,

[0105] where , is the number of hidden neurons, is the center vector of the th hidden neuron, is the Euclidean distance between the two, is the radius of the th neuron.

[0106] The output of the th RBF neural network is:

[0107] ,

[0108] where , is the th connection weight between the hidden neuron and the output layer.

[0109] Use the gradient descent method to obtain the weight update rule:

[0110] ,

[0111] .

[0112] Thus, the RBF neural network controller, which is also the specific expression of the motor input signal difference, is:

[0113] ,

[0114] where is the motor input signal difference, is the set of connection weights between each hidden neuron and the output layer, , is the set of output values of the RBF neural network hidden neurons, , is the number of output layers.

[0115] In step S106, if the vehicle meets the tracking error threshold condition or the event trigger condition, update the input signal of the vehicle motor according to the difference of the motor input signals.

[0116] Specifically, this embodiment adopts an event-triggered control mechanism to minimize the number of control executions and ensure that the tracking error converges within a satisfactory bound. Among them, with the help of Ackermann vehicle kinematics, the expected motor output signal is obtained according to the vehicle's three-axis motion speed. . Mark the event sequence of the event trigger as , . The control action is only updated at the time points when the event is triggered, otherwise the control input remains unchanged.

[0117] When , and ,

[0118] When , and or ,

[0119] Update the input signal of the vehicle motor to:

[0120] ,

[0121] That is ;

[0122] Among them, the event trigger condition is or , is the event trigger error, , is the predicted output signal, , is the first parameter, , is the second parameter, , is the estimated pseudo partial derivative at time is the actual output signal, .

[0123] The model-free event-triggered differential control method of this embodiment includes three parts: unknown motor system dynamic linearization, dynamic linearization model parameter estimation, and event-triggered control mechanism design. Among them

[0124] (1) The model-free adaptive event-triggered control proposed in this embodiment is directly applied to the unknown motor system of an Ackermann vehicle. The event-triggering condition is calculated only using the input / output (I / O) data during the operation of the motor system. Therefore, the proposed method is data-based and does not rely on any explicit motor system model information.

[0125] (2) The control method proposed in this embodiment can reduce the execution times of the controller and save computational costs while ensuring system stability and control performance.

[0126] (3) The offline parameter identification algorithm is integrated into the event-triggered model-free adaptive control method, and the obtained data model is continuously updated online using input-output data, thereby adapting to changes in system dynamics and disturbances. This improves vehicle stability.

[0127] It should be understood that although the steps in the attached Figure 1 flowchart are shown in sequence according to the arrow directions, these steps do not necessarily have to be executed in the order indicated by the arrows. Unless explicitly stated in this document, there is no strict order restriction for the execution of these steps, and these steps can be executed in other orders. Moreover, Figure 1 at least a part of the steps in the attachment can include multiple sub-steps or sub-phases. These sub-steps or phases do not necessarily have to be completed at the same time, but can be executed at different times. The execution order of these sub-steps or phases does not necessarily have to be sequential either, but can be executed alternately or in rotation with at least a part of other steps or sub-steps or phases of other steps.

[0128] In the above embodiments of the present invention, the model-free event differential control method is described in detail. The above method disclosed in the present invention can be implemented by various forms of devices. Therefore, the present invention also discloses a model-free event differential control device. Combining the attachment Figure 4 below, specific embodiments are given for detailed description.

[0129] A signal acquisition module 301, configured to acquire the desired signal, real-time input signal, and real-time output signal of the vehicle;

[0130] An input-output model construction module 302, configured to construct an input-output dynamic model of the motor system according to the historical output signal and historical input signal of the vehicle motor system;

[0131] A model conversion module 303, configured to obtain a fully formatted dynamic linearized model of the vehicle motor system according to the input-output dynamic model;

[0132] A pseudo partial derivative estimation module 304, configured to obtain an estimated pseudo partial derivative according to the historical input signal, historical output signal, and initial value of the pseudo partial derivative of the vehicle motor system;

[0133] An input signal difference determination module 305, configured to determine a following error according to the desired signal and the real-time output signal, and determine the input signal difference of the motor according to the following error and the pseudo-gradient in combination with an RBF neural network controller;

[0134] An input signal update module 306, configured to update the input signal of the vehicle motor according to the motor input signal difference if the vehicle meets the tracking error threshold condition or the event trigger condition.

[0135] Further, the pseudo-partial derivative estimation module 304 includes:

[0136] A parameter setting unit 401, configured to make , set a preset data length and an initial value of the pseudo-partial derivative ;

[0137] A historical signal acquisition unit 402, configured to acquire the historical input signal of the vehicle motor system and the historical output signal at the corresponding moment;

[0138] A pseudo-partial derivative estimation unit 403, configured to input the initial value of the pseudo-partial derivative, the historical input signal of the vehicle motor system, and the historical output signal at the corresponding moment into the specific expression for estimating the pseudo-partial derivative , and obtain an estimated pseudo-partial derivative ;

[0139] An estimated pseudo-partial derivative correction and output unit 404, configured to calculate a predicted output value of the Ackermann vehicle motor system at the next moment based on the estimated pseudo-partial derivative, and obtain a prediction error according to the predicted output value and the actual output value at the next moment , if the absolute value of the prediction error is less than a preset threshold, output the estimated pseudo-partial derivative; otherwise, make updated to , and repeat steps S202 - S204 until the absolute value of the prediction error is less than the preset threshold.

[0140] For the model-free event differential control device, reference can be made to the above limitations on the method for all, and details will not be repeated here. Each module in the above device can be implemented in whole or in part through software, hardware, and their combination. Each of the above modules can be embedded in the processor of the terminal device in hardware form or be independent of it, or can be stored in the memory of the terminal device in software form so that the processor can call and execute the operations corresponding to each of the above modules.

[0141] In one embodiment, the present invention further provides a computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, the steps of the above model-free event differential control method are implemented.

[0142] The computer-readable storage medium may be an electronic memory such as a flash memory, EEPROM (Electrically Erasable Programmable Read-Only Memory), EPROM (Erasable Programmable Read-Only Memory), a hard disk, or a ROM. Optionally, the computer-readable storage medium includes a non-transitory computer-readable storage medium. The computer-readable storage medium has a storage space for program codes for performing any method steps in the above method. These program codes can be read from or written into one or more computer program products, and the program codes can be compressed in a suitable form.

[0143] In one embodiment, the present invention provides a computer device, including a memory and a processor, where the memory stores a computer program, and when the processor executes the computer program, it executes the above model-free event differential control method.

[0144] The computer device includes a memory, a processor, and one or more computer programs, where one or more computer programs can be stored in the memory and configured to be executed by one or more processors, and one or more application programs are configured to execute the above model-free event differential control method.

[0145] The processor may include one or more processing cores. The processor connects various parts within the entire computer device using various interfaces and lines, and by running or executing instructions, programs, code sets, or instruction sets stored in the memory, and by calling data stored in the memory, it executes various functions of the computer device and processes data. Optionally, the processor may be implemented in at least one hardware form of Digital Signal Processing (DSP), Field-Programmable Gate Array (FPGA), or Programmable Logic Array (PLA). The processor may integrate one or a combination of a Central Processing Unit (CPU), a Graphics Processing Unit (GPU) for reporting and verifying buried point data, and a modem, etc. Among them, the CPU mainly processes the operating system, user interface, and application programs, etc.; the GPU is responsible for rendering and drawing display content; the modem is used for processing wireless communication. It can be understood that the above modem may not be integrated into the processor and may be implemented separately by a communication chip.

[0146] The memory may include a Random Access Memory (RAM), and may also include a Read-Only Memory. The memory can be used to store instructions, programs, code, code sets or instruction sets. The memory may include a program storage area and a data storage area. Among them, the program storage area can store instructions for implementing the operating system, instructions for implementing at least one function (such as touch function, sound playback function, image playback function, etc.), instructions for implementing the above-mentioned various method embodiments, etc. The data storage area can also store data created during the use of the terminal device, etc.

[0147] The above embodiments are only used to illustrate the technical solutions of the present invention, rather than limiting them; although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions described in the foregoing embodiments, or perform equivalent replacements for some of the technical features; and these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A model-free event differential control method, characterized in that Including: Obtain the desired signal, real-time input signal, and real-time output signal of the vehicle; Construct an input-output dynamic model of the motor system according to the historical output signal and historical input signal of the vehicle motor system; Obtain the full-format dynamic linearization model of the vehicle motor system according to the input-output dynamic model; Obtain the predicted pseudo partial derivative according to the historical input signal, historical output signal, and initial value of the pseudo partial derivative of the vehicle motor system; Determine the following error according to the desired signal and real-time output signal, and determine the difference in the motor input signal according to the following error and pseudo gradient combined with the RBF neural network controller; If the vehicle meets the tracking error threshold condition or event trigger condition, update the input signal of the vehicle motor according to the difference in the motor input signal.

2. The model-free event differential control method according to claim 1, characterized in that, The specific expression for constructing the input-output dynamic model of the motor system according to the historical output signal and historical input signal of the vehicle motor system is: ， Among them, the specific expression of the historical output signal of the vehicle motor system is , is the actual rotational speed of the left rear wheel of the vehicle at time is the actual rotational speed of the right rear wheel of the vehicle at time is the actual steering angle of the steering servo of the vehicle at time; the specific expression of the historical input signal of the vehicle motor system is , is the PWM control signal of the left rear wheel of the vehicle motor system at time is the PWM control signal of the right rear wheel of the vehicle motor system at time is the PWM control signal of the steering servo of the vehicle motor system at time; is an unknown continuously differentiable function, is the order of the first positive definite system, is the order of the second positive definite system.

3. The model-free event differential control method according to claim 2, characterized in that, The specific expression for the full-format dynamic linearization model is: ， Among them, is the change in the input and output of the vehicle motor system at a moment, , is the pseudo-partial derivative of the vehicle motor system at a moment, .

4. The model-free event differential control method according to claim 3, wherein The obtaining the predicted pseudo partial derivative according to the historical input signal, historical output signal, and initial value of the pseudo partial derivative of the vehicle motor system includes: Step S201, let , set the preset data length and the initial value of the pseudo partial derivative ; Step S202, obtain the historical input signal of the vehicle motor system and the historical output signal at the corresponding moment; Step S203, input the initial value of the pseudo partial derivative, the historical input signal of the vehicle motor system, and the historical output signal at the corresponding moment into the specific expression for estimating the pseudo partial derivative , and obtain the estimated pseudo partial derivative ; Step S204: Calculate the output prediction value of the Ackermann vehicle motor system at the next moment based on the estimated pseudo partial derivative, and obtain the prediction error according to the output prediction value and the actual output value at the next moment. If the absolute value of the prediction error is less than the preset threshold, output the estimated pseudo partial derivative; otherwise, Update to Repeat steps S202 - S204 until the absolute value of the prediction error is less than the preset threshold.

5. The model-free event differential control method according to claim 4, wherein The specific expression for the difference in the motor input signal is: ， Among them, is the difference value of the motor input signal, is the set of connection weights between each hidden neuron and the output layer, , is the set of output values of the hidden neurons of the RBF neural network, , is the number of output layers.

6. The model-free event differential control method according to claim 5, wherein The specific operation of updating the input signal of the vehicle motor according to the difference in the motor input signal if the vehicle meets the tracking error threshold condition or event trigger condition is: When and satisfying , When and satisfying or , Update the input signal of the vehicle motor to: ； Among them, the event trigger condition is or , is the event trigger error, , is the predicted output signal, , is the first parameter, , is the second parameter, , is the estimated pseudo partial derivative at the moment, is the actual output signal, .

7. A model-free event differential control device, characterized in that, Including: A signal acquisition module for obtaining the desired signal, real-time input signal, and real-time output signal of the vehicle; An input-output model construction module for constructing an input-output dynamic model of the motor system according to the historical output signal and historical input signal of the vehicle motor system; A model conversion module for obtaining the full-format dynamic linearization model of the vehicle motor system according to the input-output dynamic model; A pseudo partial derivative prediction module for obtaining the predicted pseudo partial derivative according to the historical input signal, historical output signal, and initial value of the pseudo partial derivative of the vehicle motor system; An input signal difference determination module for determining the following error according to the desired signal and real-time output signal, and determining the difference in the motor input signal according to the following error and pseudo gradient combined with the RBF neural network controller; An input signal update module for updating the input signal of the vehicle motor according to the difference in the motor input signal if the vehicle meets the tracking error threshold condition or event trigger condition.

8. The model-free event differential control device according to claim 7, characterized in that, The pseudo partial derivative prediction module includes: A parameter setting unit for making , setting a preset data length and an initial value of a pseudo partial derivative ; A historical signal acquisition unit for obtaining the historical input signal of the vehicle motor system and the historical output signal at the corresponding moment; The pseudo partial derivative estimation unit is configured to input the initial value of the pseudo partial derivative, the historical input signal of the vehicle motor system, and the historical output signal at the corresponding moment into the specific expression for estimating the pseudo partial derivative , and obtain the estimated pseudo partial derivative ; The predicted pseudo-derivative correction output unit is used to calculate the output prediction value of the Ackermann vehicle motor system at the next moment based on the predicted pseudo-derivative, and obtain the prediction error according to the output prediction value and the actual output value at the next moment , if the absolute value of the prediction error is less than the preset threshold, the predicted pseudo-derivative is output; otherwise, is updated to , and steps S202 - S204 are repeatedly executed until the absolute value of the prediction error is less than the preset threshold.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the steps of the model-free event differential speed control method according to any one of claims 1-6.

10. A computer device, comprising a memory and a processor, the memory storing a computer program, characterized in that, When the processor executes the computer program, it executes the model-free event differential speed control method according to any one of claims 1-6.

Images