A vector steering wheel cooperative control method and system with adaptive real-time optimization

Abstract

The present application relates to the field of steering wheel control technology, and provides an adaptive real-time optimized vector steering wheel collaborative control method and system. The method includes: obtaining the initial state information of the steering wheel; analyzing the initial state information to obtain the initial state; reading the state factor set, activating the observer for dynamic monitoring, and obtaining the real-time factor set; inputting the real-time factor set and the initial state into the model to obtain the predicted state; generating a collaborative control strategy for multiple steering wheel modules, and simulating the control of the steering wheel under the predicted state based on the strategy to obtain a simulation record; when the simulation record meets the constraints, determining the optimal strategy; and controlling the steering wheel according to the optimal strategy. The present application solves the technical problems of the existing steering wheel control in high dynamic loads and complex environments, such as delayed response, low control accuracy, and poor adaptability, and realizes the technical effect of quickly adjusting the steering wheel control strategy under variable operating conditions and load changes, and improving the response speed, stability, and accuracy of the steering wheel control.

Description

Technical Field

[0001] The present application relates to the technical field of steering wheel control, and in particular to an adaptive real-time optimized vector steering wheel collaborative control method and system. Background Art

[0002] With the widespread application of robotics, automated vehicles, and intelligent mobile platforms, the requirements for their motion control systems are gradually increasing. Especially in complex and changeable working environments, such as obstacle environments or high-load conditions, traditional drive systems cannot meet the high requirements for accuracy, flexibility, response speed, and adaptability. Therefore, the existing vector steering wheel system is widely used in robots and automated vehicles with high mobility requirements, such as racing robots, high-load transport vehicles, etc., because it can provide a full range of mobility and achieve precise steering and positioning. However, although the vector steering wheel system has good flexibility and agility, it still has certain limitations when facing high-frequency operations and dynamic load changes, mainly manifested in slow control response, poor system stability, low energy efficiency, and poor adaptability to external environmental changes. These shortcomings not only affect work efficiency, but also reduce reliability and durability in complex environments. Summary of the invention

[0003] The present application provides an adaptive real-time optimized vector steering wheel collaborative control method and system, aiming to solve the technical problems of response delay, low control accuracy and poor adaptability of existing steering wheel control under high dynamic loads and complex environments.

[0004] In view of the above problems, the present application provides an adaptive real-time optimized vector steering wheel collaborative control method and system.

[0005] The first aspect disclosed in the present application provides an adaptive real-time optimized vector steering wheel cooperative control method, wherein the adaptive real-time optimized vector steering wheel cooperative control method is applied to an adaptive real-time optimized vector steering wheel cooperative control system, wherein the adaptive real-time optimized vector steering wheel cooperative control system includes multiple steering wheel modules, wherein the adaptive real-time optimized vector steering wheel cooperative control method includes: obtaining initial state information of the vector steering wheel, wherein the initial state information includes steering wheel structure information and steering wheel load information; analyzing the steering wheel structure information and the steering wheel load information to obtain the initial dynamic state of the vector steering wheel; reading a set of dynamic state factors, and activating a state observer to obtain the initial state information of the vector steering wheel; Dynamically monitor the dynamic state factor set of the vector steering wheel to obtain a real-time factor parameter set; use the real-time factor parameter set and the initial dynamic state as input information of a state prediction model, and obtain output information through the state prediction model, wherein the output information includes the predicted dynamic state of the vector steering wheel; generate a collaborative control strategy for the multiple steering wheel modules, and based on the collaborative control strategy, simulate the control of the vector steering wheel under the predicted dynamic state to obtain a simulated control record; when the simulated control record meets the predetermined control constraints, use the collaborative control strategy as the optimal strategy; and perform adaptive real-time control on the vector steering wheel according to the optimal strategy.

[0006] Another aspect disclosed in the present application provides an adaptive real-time optimized vector steering wheel cooperative control system for implementing the adaptive real-time optimized vector steering wheel cooperative control method. The adaptive real-time optimized vector steering wheel cooperative control system includes multiple steering wheel modules, and the system also includes: an initial state information acquisition component: acquiring the initial state information of the vector steering wheel, the initial state information including the steering wheel structure information and the steering wheel load information; an initial state information analysis component: analyzing the steering wheel structure information and the steering wheel load information to obtain the initial dynamic state of the vector steering wheel; a dynamic monitoring component: reading a dynamic state factor set, and activating a state observer to dynamically monitor the dynamic state factor set of the vector steering wheel. Monitoring to obtain a real-time factor parameter set; a state prediction component: using the real-time factor parameter set and the initial dynamic state as input information of a state prediction model, and obtaining output information through the state prediction model, wherein the output information includes the predicted dynamic state of the vector steering wheel; a simulation control component: generating a collaborative control strategy for the multiple steering wheel modules, and based on the collaborative control strategy, performing simulated control on the vector steering wheel under the predicted dynamic state to obtain a simulated control record; an optimal strategy generation component: when the simulated control record satisfies a predetermined control constraint, taking the collaborative control strategy as the optimal strategy; an adaptive real-time control component: performing adaptive real-time control on the vector steering wheel according to the optimal strategy.

[0007] One or more technical solutions provided in this application have at least the following technical effects or advantages:

[0008] The above-mentioned adaptive real-time optimized vector steering wheel collaborative control method first obtains the initial state information of the vector steering wheel, which includes the structural characteristics and load conditions of the steering wheel. By analyzing these initial data, the initial dynamic state of the vector steering wheel can be inferred, laying the foundation for the subsequent control process; then, the active mechanical state factor set is read, and these factors are dynamically monitored by the state observer to obtain a new factor parameter set in real time. This dynamic monitoring process can timely reflect the changes of the vector steering wheel during operation and ensure that the optimal performance is maintained under various environmental conditions; then, the real-time factor parameter set is input into the state prediction model together with the initial dynamic state to predict the The future dynamic state of the vector steering wheel; after obtaining the predicted dynamic state, a collaborative control strategy for multiple steering wheel modules is generated, and the predicted state of the vector steering wheel is simulated and controlled based on the strategy. Through the simulated control record, the effects of different control schemes can be evaluated to ensure that the selected strategy can meet the predetermined control constraints; only when the simulated control record meets the preset performance standards will the strategy be selected as the optimal strategy. This process ensures the scientificity and rationality of the strategy selection; finally, based on the optimal strategy, the vector steering wheel will perform adaptive real-time control to ensure that the vector steering wheel always performs its tasks in the best way. This not only improves the operating accuracy, but also enhances the ability to adapt to sudden changes.

[0009] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, it can be implemented in accordance with the contents of the specification. In order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are listed below. BRIEF DESCRIPTION OF THE DRAWINGS

[0010] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0011] Figure 1 The figure is a flow chart of an adaptive real-time optimized vector steering wheel cooperative control method in one embodiment.

[0012] Figure 2 This is an architecture diagram of an adaptive real-time optimized vector steering wheel collaborative control system in one embodiment.

[0013] Explanation of the reference numerals: initial state information acquisition component 1, initial state information analysis component 2, dynamic monitoring component 3, state prediction component 4, simulation control component 5, optimal strategy generation component 6, adaptive real-time control component 7. DETAILED DESCRIPTION

[0014] The embodiments of the present application provide an adaptive real-time optimized vector steering wheel collaborative control method and system to solve the technical problems of response delay, low control accuracy and poor adaptability of existing steering wheel control under high dynamic loads and complex environments.

[0015] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of this application.

[0016] It should be noted that the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions. For example, a process, method, system, product or server that includes a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or modules that are not explicitly listed or inherent to these processes, methods, products or devices.

[0017] Embodiment 1, as Figure 1 As shown, the present application provides an adaptive real-time optimized vector steering wheel collaborative control method, the adaptive real-time optimized vector steering wheel collaborative control method is applied to an adaptive real-time optimized vector steering wheel collaborative control system, the adaptive real-time optimized vector steering wheel collaborative control system includes multiple steering wheel modules, wherein the adaptive real-time optimized vector steering wheel collaborative control method includes:

[0018] The initial state information of the vector steering wheel is obtained, wherein the initial state information includes the steering wheel structure information and the steering wheel load information.

[0019] An adaptive real-time optimized vector steering wheel cooperative control method is applied to an adaptive real-time optimized vector steering wheel cooperative control system. The system includes multiple steering wheel modules, each steering wheel module includes an integrated hub unit, which rotates around a central axis. The hub unit consists of a ring gear and multiple pinions, which are connected to the motor shaft. The motor drives the torque to the ring gear through a gear set, thereby driving the hub to rotate. The relative speed difference of the motor is used to control the steering of the steering wheel, and the average speed of the motor determines the forward speed of the steering wheel. This design not only improves the control accuracy, but also optimizes the energy utilization efficiency.

[0020] In the embodiment of the present application, when obtaining the initial state information of the vector steering wheel, it is necessary to record the structural information and load information of the vector steering wheel in detail. These two parts are the basis for subsequent control optimization; the steering wheel structural information of the vector steering wheel mainly includes the steering wheel diameter, steering wheel material, steering wheel structure and steering wheel mass. The steering wheel diameter is a key parameter affecting the rotational inertia and torque output of the vector steering wheel. Vector steering wheels with larger diameters usually have larger rotational inertia, which will affect the acceleration and deceleration capabilities of the vector steering wheel, and thus affect the response speed. The material (such as aluminum alloy, steel, plastic, etc.) and structural design (such as the shape of the hub, strength distribution, etc.) of the vector steering wheel will also have an important influence on the stiffness, strength and durability of the steering wheel. The elastic modulus, tensile strength and fatigue resistance of different materials determine the ability of the steering wheel to withstand torque during operation. The mass of the vector steering wheel is another key factor affecting its inertial force. The larger the mass of the vector steering wheel, the greater its inertial force, which means that more torque is required to change its rotation state; the steering wheel load information of the vector steering wheel is a description of the vector steering wheel load information. The present invention discloses the external forces to which the vector steering wheel is subjected during operation, including the load and the dynamic change of the load. The size and distribution of the steering wheel load have a direct impact on the stress state and power demand of the vector steering wheel. In practical applications, the load may be static (such as the dead weight of a robot or automated vehicle, additional cargo or load). By measuring the total mass of the vehicle or robot and combining the load distribution of each steering wheel, the load data borne by each steering wheel can be obtained. The load is not only affected by static factors, but also dynamic loads (such as the inertial force generated during acceleration, deceleration or steering) need to be considered. Whenever the motion state changes, the inertial force will change with the change of speed and acceleration. The dynamic load is calculated by real-time acquisition of acceleration, angular velocity and other data and combining the Newton-Euler method, the D'Alembert method and the like. By sorting out the obtained structural information and load information of the vector steering wheel, the initial state information of the vector steering wheel can be obtained. This information can accurately evaluate the initial dynamic state of the steering wheel and ensure that the vector steering wheel can operate stably and accurately under various working conditions.

[0021] The initial dynamic state of the vector steering wheel is obtained by analyzing the steering wheel structure information and the steering wheel load information.

[0022] In one embodiment, after obtaining the steering wheel structure information and the steering wheel load information, the system terminal analyzes and extracts the steering wheel structure information, obtains the moment of inertia, force state (static load, dynamic load, friction), motion state (acceleration, angular velocity, speed), etc. of the vector steering wheel, and summarizes these data to form the initial dynamic state of the vector steering wheel; for example, if the vector steering wheel is a disc steering wheel, the steering wheel radius r and the steering wheel mass m in the steering wheel structure information can be combined to obtain the initial dynamic state of the vector steering wheel. Calculate the moment of inertia of the vector steering wheel. If the vector steering wheel is a ring-shaped steering wheel, the inner radius r of the steering wheel in the steering wheel structure information is 1 , outer radius r 2 Combined with the steering wheel mass m, The moment of inertia of the vector steering wheel is calculated. The friction force can be obtained by multiplying the static load in the steering wheel load information with the friction coefficient of the steering wheel material in the steering wheel structure information. These initial dynamic states provide key inputs for subsequent real-time monitoring, dynamic adjustment, and optimization of control strategies, ensuring that the steering wheel can respond to external inputs and load changes in the best way in a complex environment.

[0023] The dynamic state factor set is read, and the state observer is activated to dynamically monitor the dynamic state factor set of the vector steering wheel to obtain a real-time factor parameter set.

[0024] In one embodiment, during this process, the system terminal reads a pre-assembled dynamic state factor set, which contains multiple parameters related to the environment and operating conditions, which can significantly affect the motion performance of the vector steering wheel, such as terrain features, ground roughness, wind speed, temperature, etc.; once the dynamic state factor set is read, the system terminal will activate the state observer, monitor these external environmental factors in real time, and collect the corresponding parameters in the dynamic state factor set to obtain a real-time factor parameter set. This real-time factor parameter set reflects the actual impact of the external environment on the vector steering wheel. Through these parameters, the control strategy of the vector steering wheel can be dynamically adjusted when the external environment changes to ensure motion accuracy, stability and flexibility.

[0025] Further, the present application provides a method of reading a dynamic state factor set, and activating a state observer to dynamically monitor the dynamic state factor set of the vector steering wheel to obtain a real-time factor parameter set, including:

[0026] Acquire the historical control record of the vector steering wheel; extract the first historical record in the historical control record, wherein the first historical record includes the first historical environmental information, the first historical predetermined trajectory and the first historical actual trajectory of the vector steering wheel; compare and analyze the first historical predetermined trajectory and the first historical actual trajectory to obtain the first historical trajectory accuracy, and use the first historical trajectory accuracy as the dependent variable; construct an independent variable based on the first historical environmental information, and perform correlation analysis with the dependent variable to obtain a correlation result; analyze the correlation result and construct the dynamic state factor set.

[0027] Preferably, the system terminal extracts control records from the historical log of the vector steering wheel to obtain the historical control record of the vector steering wheel, which contains information such as the past control operations of the vector steering wheel, external environmental data, and corresponding motion trajectories; then, the steering wheel control information within a specific moment or period is randomly extracted from the historical control record as a first historical record, and the first historical record includes first historical environmental information, a first historical predetermined trajectory, and a first historical actual trajectory, wherein the first historical environmental information refers to relevant data of the environment in which the vector steering wheel is located at that moment or period (such as ground conditions, climatic conditions, obstacle distribution, etc.), and these data reflect the environmental conditions at that time, the first historical predetermined trajectory refers to the ideal trajectory that the vector steering wheel should complete according to the control command, and the first historical actual trajectory refers to the actual path taken by the steering wheel during actual execution under external interference; then, the first historical predetermined trajectory and the first historical actual trajectory are compared, and the trajectory accuracy is obtained by calculating the difference in the trajectories, and the trajectory accuracy reflects the accuracy of the vector steering wheel in execution. The deviation between the expected trajectory and the expected trajectory during the execution process is calculated as follows: the trajectory deviation between the predetermined position and the actual position of each pair of data points is calculated by Euclidean distance (or maximum deviation, etc.), and then all the calculated trajectory deviations are accumulated and averaged to obtain the trajectory accuracy of the first historical predetermined trajectory and the first historical actual trajectory; after obtaining the trajectory accuracy and taking it as the dependent variable, the system terminal constructs independent variables related to the trajectory accuracy. The independent variables are usually external factors that affect the execution of the trajectory. These factors are extracted from the first historical environmental information, such as ground roughness, wind speed, wind direction, etc., and then the correlation between each variable and the dependent variable is analyzed by statistical methods (such as Pearson correlation coefficient, regression analysis, etc.); then, according to the correlation analysis results, combined with the correlation threshold (determined according to historical experience and expert advice), which independent variables have a significant impact on the dependent variable are screened out, and a set of dynamic state factors is constructed based on the screening results. These factors can significantly affect the movement and control of the vector steering wheel, providing data support and optimization basis for subsequent adaptive real-time control.

[0028] Further, the present application provides a method for analyzing the correlation results and forming the kinetic state factor set, including:

[0029] Obtain a type of state factor; obtain a type of state factor; the type of state factor and the type of state factor constitute the dynamic state factor set; wherein, obtaining the type of state factor includes: forming a type of environmental feature; screening the first historical environmental information based on the type of environmental feature, and forming a type of feature parameter set, and using the type of feature parameter set as a type of independent variable; screening the type of environmental feature according to the correlation analysis between the type of independent variable and the dependent variable, and obtaining a type of significant environmental feature; using the type of significant environmental feature as the type of state factor; wherein, obtaining the type of state factor includes: forming a type of environmental feature; screening the first historical environmental information based on the type of environmental feature, and forming a type of feature parameter set, and using the type of feature parameter set as a type of independent variable; screening the type of environmental feature according to the correlation analysis between the type of independent variable and the dependent variable, and obtaining a type of significant environmental feature; using the type of significant environmental feature as the type of state factor.

[0030] Optionally, the dynamic state factor set is obtained by merging a class of state factors and a class of state factors. Both of these class of state factors are obtained by performing a correlation analysis on the first historical environmental information and then combining the correlation threshold screening. Specifically, for the class of state factors, the system terminal first forms a class of environmental features. The class of environmental features includes static factors such as ground characteristics (such as hardness, slope, roughness, ground type), terrain characteristics (such as obstacle distribution, terrain flatness, terrain complexity), etc. These static factors may affect the motion performance and control accuracy of the steering wheel; then, data corresponding to the class of environmental features are screened out from the first historical environmental information to form a set of class of feature parameter sets, and then the same operation is performed on other records in the historical control records in the same manner as described above to obtain multiple sets of class of feature parameter sets. These class of feature parameter sets will be used as a class of independent variables and are subjected to correlation analysis with the dependent variable (also including the trajectory accuracy of other records in the historical control records) to screen out environmental features that have a significant impact on the dependent variable; the system terminal calculates the correlation between the class of independent variables and the dependent variable based on the Pearson correlation coefficient to obtain each feature in the class of independent variables. (such as hardness, slope, etc.); then, the correlation coefficient of each feature is compared with the correlation threshold, and features greater than or equal to the correlation threshold are screened out, and these screened features are selected as a class of significant environmental features into a class of state factors, and these class one state factors include static environmental factors that can significantly affect the control and motion accuracy of the vector steering wheel, such as hardness, slope, roughness, etc.; based on the same method, the system terminal constructs a second class of environmental features, and this second class of environmental features includes dynamic factors such as wind speed, wind force, wind direction, temperature, humidity, and light. Changes in these dynamic factors may have an immediate impact on the movement of the vector steering wheel; after constructing the second class of environmental features, the system terminal again screens the first historical environmental information, determines the second class of feature parameter sets and uses them as the second class of independent variables, performs correlation analysis with the dependent variable, and then based on the analysis results, combines the correlation threshold to screen out the second class of state factors; the system terminal combines the first class of state factors and the second class of state factors into a complete set of dynamic state factors for subsequent state prediction to ensure that the vector steering wheel can cope with complex external environmental changes and maintain precise control and stable movement.

[0031] The real-time factor parameter set and the initial dynamic state are used as input information of a state prediction model, and output information is obtained through the state prediction model, wherein the output information includes the predicted dynamic state of the vector steering wheel.

[0032] In one embodiment, the system terminal passes the real-time factor parameter set and the initial dynamic state as input information to the state prediction model. The goal of this model is to predict the future motion state based on the known current environment and the initial state of the vector steering wheel. Through the operation of the state factor structure diagram in the state prediction model, the system terminal will obtain an output information based on the input environmental real-time factor parameter set and the initial dynamic state. This output information includes predicted dynamic state, such as predicted moment of inertia, predicted friction, predicted acceleration, predicted angular velocity, etc., to help the system terminal understand the future motion trend of the vector steering wheel in advance, which is crucial for adjusting the control strategy and optimizing the response speed and accuracy of the vector steering wheel.

[0033] Further, the present application provides using the real-time factor parameter set and the initial dynamic state as input information of a state prediction model, and obtaining output information through the state prediction model, wherein the output information includes the predicted dynamic state of the vector steering wheel, including:

[0034] Extract the state factor structure diagram in the state prediction model; perform state influence evaluation on the real-time factor parameter set in combination with the state factor structure diagram to obtain a real-time influence coefficient; adjust the initial dynamic state using the real-time influence coefficient as a weight to obtain the predicted dynamic state.

[0035] Preferably, after the real-time factor parameter set and the initial dynamic state are input into the state prediction model, the system terminal activates the state prediction model. The state prediction model is a graphical model structure, which includes a state factor structure diagram. After activating the state prediction model, the system terminal extracts the state factor structure diagram, and based on the state factor structure diagram, performs state influence evaluation on the real-time factor parameter set by matrix calculation, and calculates the real-time influence coefficient to reflect the actual contribution of these factors to the prediction result; then, the real-time influence coefficient is weightedly calculated with the parameters in the initial dynamic state to obtain the predicted dynamic state, so that the system terminal can adapt and adjust the control strategy in real time in a changing environment to ensure that the vector steering wheel accurately and stably performs the predetermined task.

[0036] Furthermore, the present application provides a method for extracting a state factor structure diagram in the state prediction model, including:

[0037] The state influence is taken as the first-level vertex, and each state factor in the dynamic state factor set is recorded as the second-level vertex; any state factor in the dynamic state factor set is extracted, and the expert decision method is used to obtain the arbitrary influence index of the arbitrary state factor; the association relationship between the second-level vertex and the first-level vertex is established according to the arbitrary influence index to obtain the state factor structure diagram.

[0038] Preferably, when constructing a state prediction model, it is first necessary to construct a state factor structure diagram. The system terminal uses the state influence as the first-level vertex in the state factor structure diagram, and each state factor in the dynamic state factor set (such as ground characteristics, environmental conditions, etc.) is used as a second-level vertex. The first-level vertex (state influence) represents the main factor affecting the control and movement of the vector steering wheel. It is usually a comprehensive influence indicator or decision result, which reflects the overall influence of all environmental and internal factors on the steering wheel state. The second-level vertex (state factor) represents the specific dynamic state factor, for example, ground hardness, slope, wind speed, temperature, etc., are all specific variables that affect the steering wheel movement accuracy. Each second-level vertex represents A factor, and the relationship between them and the first-level vertices describes the influence of each factor on the steering wheel control; then, the system terminal randomly extracts the dynamic state factor from the dynamic state factor set, obtains an arbitrary state factor, and obtains an arbitrary influence index of this arbitrary state factor according to the expert decision-making method, where the expert decision-making method is a way to evaluate the influence of factors based on experience, knowledge and judgment. In this process, experts judge the influence of a certain state factor (such as the ground friction coefficient) on the state of the vector steering wheel based on actual application scenarios and experience. The system terminal collects the opinions of experts in the field, calculates the mean of the opinions of experts in the field, and determines the influence index of each state factor. For example, experts can give a score value (usually a value between 0 and 1, 0 means no effect, 1 means great effect) according to how the change of ground friction coefficient affects the traction and control accuracy of the steering wheel. Then, according to the influence index of each state factor obtained, the relationship between the secondary vertex and the primary vertex is established, that is, the influence index of each state factor is used as the edge between the secondary vertex and the primary vertex, so as to construct a state factor structure diagram. The system terminal constructs a state prediction model by connecting the state factor structure diagram with the input nodes and the output nodes, laying the foundation for the subsequent prediction of the dynamic state of the vector steering wheel.

[0039] Furthermore, the present application provides a method for evaluating the state influence of the real-time factor parameter set in combination with the state factor structure diagram to obtain a real-time influence coefficient, including:

[0040] The diagonal matrix of the state factor structure diagram is obtained by analysis; the adjacency matrix of the state factor structure diagram is obtained by analysis; the Laplace matrix of the state factor structure diagram is calculated according to the diagonal matrix and the adjacency matrix; the Laplace matrix is ​​subjected to eigendecomposition processing in combination with the real-time factor parameter set to obtain the eigenvalue of the state factor structure; the eigenvalue of the state factor structure is recorded as the real-time influence coefficient.

[0041] Optionally, after extracting the state factor structure diagram, the system terminal extracts the influence index in the state factor structure diagram and constructs a diagonal matrix. The diagonal matrix is ​​a square matrix, and the elements on its diagonal represent the influence of each state factor. Any secondary vertices in the state factor structure diagram are combined in pairs, and the influence degree between each combination is evaluated by the expert decision method to determine the influence index between each combination. Then, the influence index between these combinations is used to construct an adjacency matrix. The adjacency matrix describes the influence between two nodes (state factors). Subsequently, the diagonal matrix and the adjacency matrix are difference-calculated to obtain the Laplace matrix. The interdependence and relationship between nodes in the mapping diagram; then, the eigenvalue decomposition method is used to decompose the eigenvalue of each node from the Laplace matrix, that is, the decomposition is performed by calculating det(L-λI)=0, where L is the Laplace matrix to be decomposed, λ is the eigenvalue obtained by decomposition, and I is the unit matrix (a square matrix, whose characteristic is that the elements on the main diagonal are all 1, and the elements in other positions are 0); then, the eigenvalue of each node is weighted averaged by the influence index of each node to obtain the eigenvalue of the state factor structure, and it is recorded as the real-time influence coefficient, which is used for the subsequent adjustment of the initial dynamic state to help make more precise adjustments and optimizations.

[0042] A collaborative control strategy for the multiple steering wheel modules is generated, and based on the collaborative control strategy, the vector steering wheel under the predicted dynamic state is simulated controlled to obtain a simulated control record.

[0043] In one embodiment, the system terminal randomly generates collaborative control strategies for multiple steering wheel modules from a collaborative control strategy library to ensure that the steering wheel modules work in coordination. The core goal of collaborative control is to coordinate the control inputs (such as speed, steering angle) of all steering wheel modules to achieve the overall goal (such as path tracking, smooth motion control, etc.); then, simulation control is performed according to the collaborative control strategy and the predicted dynamic state. Simulation control is to test the effectiveness of the control strategy through a simulation platform. In this process, the system terminal inputs the predicted dynamic state and collaborative control strategy to the simulation platform. The simulation platform will perform simulation based on the input data and record the data of the simulation process to form a simulation control record, including motion trajectory, energy consumption data, control response, etc. These data provide a basis for the subsequent selection of collaborative control strategies.

[0044] When the simulated control record satisfies the predetermined control constraint, the collaborative control strategy is used as the optimal strategy; and the vector steering wheel is adaptively controlled in real time according to the optimal strategy.

[0045] In one embodiment, after obtaining the simulated control record of the collaborative control strategy, the system terminal analyzes the simulated control record to identify the energy consumption and control accuracy in the collaborative control process, and calculates the fitness of the collaborative control strategy based on the identified content, and then compares the calculation result with the predetermined control constraints to determine whether the collaborative control strategy is the current optimal strategy. If so, the vector steering wheel is adaptively controlled in real time according to this optimal strategy to ensure that the vector steering wheel can maintain stable and precise control in a changing environment. If not, a new collaborative control strategy is selected and the above process is repeated until the requirements of the predetermined control constraints are met.

[0046] Furthermore, the present application includes:

[0047] Extracting a first control record of a first steering wheel module in the simulation control record, wherein the first steering wheel module refers to any one of the multiple steering wheel modules; analyzing the first control record to obtain a first control energy consumption, and calculating a comprehensive control energy consumption based on the first control energy consumption; comparing and analyzing the simulated actual trajectory with the simulated predetermined trajectory in the simulation control record to obtain a simulation control accuracy; performing weighted calculation on the normalized comprehensive control energy consumption and the simulation control accuracy to obtain a simulation control fitness; when the simulation control fitness reaches a predetermined fitness threshold in the predetermined control constraint, the simulation control record satisfies the predetermined control constraint.

[0048] Preferably, the system terminal extracts the first control record of the first steering wheel module from the simulated control record, and the first steering wheel module refers to any one of all the steering wheel modules; then, based on the extracted first control record of the first steering wheel module, the system terminal analyzes its energy consumption performance, and extracts the first control energy consumption therefrom, for example, the power of the motor, and then summarizes the energy consumption of all the steering wheel modules to obtain a comprehensive control energy consumption; thereafter, using the same method as described above, the simulated control accuracy of the simulated actual trajectory and the simulated predetermined trajectory in the simulated control record is calculated by Euclidean distance (or maximum deviation, etc.); then, the comprehensive control energy consumption and the simulated control accuracy are normalized by the maximum and minimum method, and the comprehensive control energy consumption and the simulated control accuracy are normalized. The simulation results are standardized to the same scale to avoid a certain indicator from excessively affecting the result. Then, the normalized comprehensive control energy consumption and simulated control accuracy are weightedly calculated by the preset weights (determined according to historical experience and expert advice) to obtain the simulated control fitness. Finally, the simulated control fitness is compared with the predetermined fitness threshold in the predetermined control constraint to determine whether the current collaborative control strategy meets the requirements. If the simulated control fitness is greater than or equal to the predetermined fitness threshold, it means that the collaborative control strategy corresponding to the simulated control record meets all constraints (such as energy efficiency, control accuracy, etc.) and can be considered to be the optimal strategy that meets the requirements. If not, it is necessary to re-acquire the collaborative control strategy and simulate it until the predetermined requirements are met.

[0049] In summary, the embodiments of the present application have at least the following technical effects:

[0050] The embodiment of the present application aims to achieve precise control of the vector steering wheel by dynamically monitoring the structure and load information of the steering wheel and adjusting the real-time factor parameter set. First, the initial state information of the vector steering wheel is obtained and analyzed, and dynamic monitoring is performed in combination with the dynamic state factor set to generate a predicted dynamic state. Based on this, a collaborative control strategy for multiple steering wheel modules is formulated. Through simulation control, energy consumption and control accuracy are evaluated, and the optimal strategy is selected for adaptive real-time control to ensure efficient operation while meeting accuracy and energy efficiency requirements. These technical effects jointly solve the technical problems of the existing steering wheel control, such as response delay, low control accuracy, and poor adaptability under high dynamic loads and complex environments, and achieve the technical effect of quickly adjusting the steering wheel control strategy under variable operating conditions and load changes, and improving the response speed, stability, and accuracy of the steering wheel control.

[0051] Embodiment 2, based on the same inventive concept as the adaptive real-time optimized vector steering wheel cooperative control method in the above embodiment, Figure 2As shown, the present application provides an adaptive real-time optimized vector steering wheel cooperative control system, the adaptive real-time optimized vector steering wheel cooperative control system includes multiple steering wheel modules, and the system also includes: an initial state information acquisition component 1: acquiring the initial state information of the vector steering wheel, the initial state information including the steering wheel structure information and the steering wheel load information; an initial state information analysis component 2: analyzing the steering wheel structure information and the steering wheel load information to obtain the initial dynamic state of the vector steering wheel; a dynamic monitoring component 3: reading the dynamic state factor set, and activating the state observer to dynamically monitor the dynamic state factor set of the vector steering wheel to obtain a real-time factor parameter set; a state prediction component 3: Component 4: uses the real-time factor parameter set and the initial dynamic state as input information of the state prediction model, and obtains output information through the state prediction model, wherein the output information includes the predicted dynamic state of the vector steering wheel; simulation control component 5: generates a collaborative control strategy for the multiple steering wheel modules, and performs simulated control on the vector steering wheel under the predicted dynamic state based on the collaborative control strategy to obtain a simulated control record; optimal strategy generation component 6: when the simulated control record meets the predetermined control constraints, uses the collaborative control strategy as the optimal strategy; adaptive real-time control component 7: performs adaptive real-time control on the vector steering wheel according to the optimal strategy.

[0052] Furthermore, the dynamic monitoring component 3 is also used to execute the following method:

[0053] Acquire the historical control record of the vector steering wheel; extract the first historical record in the historical control record, wherein the first historical record includes the first historical environmental information, the first historical predetermined trajectory and the first historical actual trajectory of the vector steering wheel; compare and analyze the first historical predetermined trajectory and the first historical actual trajectory to obtain the first historical trajectory accuracy, and use the first historical trajectory accuracy as the dependent variable; construct an independent variable based on the first historical environmental information, and perform correlation analysis with the dependent variable to obtain a correlation result; analyze the correlation result and construct the dynamic state factor set.

[0054] Furthermore, the dynamic monitoring component 3 is also used to execute the following method:

[0055] Obtain a type of state factor; obtain a type of state factor; the type of state factor and the type of state factor constitute the dynamic state factor set; wherein, obtaining the type of state factor includes: forming a type of environmental feature; screening the first historical environmental information based on the type of environmental feature, and forming a type of feature parameter set, and using the type of feature parameter set as a type of independent variable; screening the type of environmental feature according to the correlation analysis between the type of independent variable and the dependent variable, and obtaining a type of significant environmental feature; using the type of significant environmental feature as the type of state factor; wherein, obtaining the type of state factor includes: forming a type of environmental feature; screening the first historical environmental information based on the type of environmental feature, and forming a type of feature parameter set, and using the type of feature parameter set as a type of independent variable; screening the type of environmental feature according to the correlation analysis between the type of independent variable and the dependent variable, and obtaining a type of significant environmental feature; using the type of significant environmental feature as the type of state factor.

[0056] Furthermore, the state prediction component 4 is also used to execute the following method:

[0057] Extract the state factor structure diagram in the state prediction model; perform state influence evaluation on the real-time factor parameter set in combination with the state factor structure diagram to obtain a real-time influence coefficient; adjust the initial dynamic state using the real-time influence coefficient as a weight to obtain the predicted dynamic state.

[0058] Furthermore, the state prediction component 4 is also used to execute the following method:

[0059] The state influence is taken as the first-level vertex, and each state factor in the dynamic state factor set is recorded as the second-level vertex; any state factor in the dynamic state factor set is extracted, and the expert decision method is used to obtain the arbitrary influence index of the arbitrary state factor; the association relationship between the second-level vertex and the first-level vertex is established according to the arbitrary influence index to obtain the state factor structure diagram.

[0060] Furthermore, the state prediction component 4 is also used to execute the following method:

[0061] The diagonal matrix of the state factor structure diagram is obtained by analysis; the adjacency matrix of the state factor structure diagram is obtained by analysis; the Laplace matrix of the state factor structure diagram is calculated according to the diagonal matrix and the adjacency matrix; the Laplace matrix is ​​subjected to eigendecomposition processing in combination with the real-time factor parameter set to obtain the eigenvalue of the state factor structure; the eigenvalue of the state factor structure is recorded as the real-time influence coefficient.

[0062] Furthermore, the optimal strategy generation component 6 is also used to execute the following method:

[0063] Extracting a first control record of a first steering wheel module in the simulation control record, wherein the first steering wheel module refers to any one of the multiple steering wheel modules; analyzing the first control record to obtain a first control energy consumption, and calculating a comprehensive control energy consumption based on the first control energy consumption; comparing and analyzing the simulated actual trajectory with the simulated predetermined trajectory in the simulation control record to obtain a simulation control accuracy; performing weighted calculation on the normalized comprehensive control energy consumption and the simulation control accuracy to obtain a simulation control fitness; when the simulation control fitness reaches a predetermined fitness threshold in the predetermined control constraint, the simulation control record satisfies the predetermined control constraint.

[0064] It should be noted that the above-mentioned sequence of the embodiments of the present application is only for description and does not represent the advantages and disadvantages of the embodiments. And the above-mentioned specific embodiments of this specification are described. The processes depicted in the accompanying drawings do not necessarily require the specific order and continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0065] The above description is only a preferred embodiment of the present application and is not intended to limit the present application. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present application shall be included in the protection scope of the present application.

[0066] This specification and drawings are merely exemplary illustrations of the present application and are deemed to cover any and all modifications, variations, combinations or equivalents within the scope of the present application. Obviously, a person skilled in the art may make various modifications and variations to the present application without departing from the scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the present application and its equivalents, the present application intends to include these modifications and variations.

Claims

1. An adaptive real-time optimized vector steering wheel cooperative control method, characterized in that: The adaptive real-time optimized vector steering wheel cooperative control method is applied to an adaptive real-time optimized vector steering wheel cooperative control system, and the adaptive real-time optimized vector steering wheel cooperative control system includes multiple steering wheel modules, wherein the adaptive real-time optimized vector steering wheel cooperative control method includes: Acquiring initial state information of the vector steering wheel, wherein the initial state information includes steering wheel structure information and steering wheel load information; Analyzing the steering wheel structure information and the steering wheel load information to obtain the initial dynamic state of the vector steering wheel; Reading a dynamic state factor set, and activating a state observer to dynamically monitor the dynamic state factor set of the vector steering wheel to obtain a real-time factor parameter set, wherein the dynamic state factor set includes a plurality of parameters related to the environment and operating conditions; Using the real-time factor parameter set and the initial dynamic state as input information of a state prediction model, and obtaining output information through the state prediction model, wherein the output information includes the predicted dynamic state of the vector steering wheel; generating a coordinated control strategy for the plurality of steering wheel modules, and performing simulated control on the vector steering wheel under the predicted dynamic state based on the coordinated control strategy to obtain a simulated control record; When the simulated control record satisfies the predetermined control constraint, taking the collaborative control strategy as the optimal strategy; The vector steering wheel is adaptively controlled in real time according to the optimal strategy.

2. The adaptive real-time optimized vector steering wheel cooperative control method according to claim 1 is characterized in that: The dynamic state factor set is read, and the state observer is activated to dynamically monitor the dynamic state factor set of the vector steering wheel to obtain a real-time factor parameter set, including: Obtaining historical control records of the vector steering wheel; Extracting a first historical record from the historical control record, the first historical record including first historical environmental information, a first historical predetermined trajectory, and a first historical actual trajectory of the vector steering wheel; Comparing and analyzing the first historical predetermined trajectory and the first historical actual trajectory to obtain the first historical trajectory accuracy, and taking the first historical trajectory accuracy as a dependent variable; Based on the first historical environmental information, an independent variable is formed, and a correlation analysis is performed with the dependent variable to obtain a correlation result; The correlation results are analyzed and the kinetic state factor set is constructed.

3. The adaptive real-time optimized vector steering wheel cooperative control method according to claim 2 is characterized in that: Analyzing the correlation results and forming the kinetic state factor set includes: Get a class of state factors; Get the second type of state factor; The first type of state factors and the second type of state factors constitute the dynamic state factor set; Among them, a type of state factor is obtained, including: Build a class of environmental features; The first historical environment information is screened based on the first type of environment characteristics to form a type of feature parameter set, and the first type of feature parameter set is used as a type of independent variable; Screening the type of environmental features according to the correlation analysis between the type of independent variables and the dependent variables to obtain a type of significant environmental features; Using the one type of significant environmental features as the one type of state factors; Among them, the second type of state factors are obtained, including: Form the second type of environmental characteristics; The first historical environmental information is screened based on the second type of environmental features, and a second type of characteristic parameter set is formed, and the second type of characteristic parameter set is used as a second type of independent variable; Screening the two types of environmental characteristics according to the correlation analysis between the two types of independent variables and the dependent variables to obtain two types of significant environmental characteristics; The two types of significant environmental characteristics are used as the two types of state factors.

4. The adaptive real-time optimized vector steering wheel cooperative control method according to claim 1 is characterized in that: The real-time factor parameter set and the initial dynamic state are used as input information of a state prediction model, and output information is obtained through the state prediction model, wherein the output information includes the predicted dynamic state of the vector steering wheel, including: Extracting a state factor structure diagram in the state prediction model; Performing a state influence evaluation on the real-time factor parameter set in combination with the state factor structure diagram to obtain a real-time influence coefficient; The initial dynamic state is adjusted using the real-time influence coefficient as a weight to obtain the predicted dynamic state.

5. The adaptive real-time optimized vector steering wheel cooperative control method according to claim 4 is characterized in that: Extracting the state factor structure diagram in the state prediction model includes: The state influence is taken as the first-level vertex, and each state factor in the dynamic state factor set is recorded as the second-level vertex; Extracting any state factor from the dynamic state factor set, and obtaining any influence index of the arbitrary state factor by using an expert decision method; The association relationship between the secondary vertex and the primary vertex is established according to the arbitrary influence index to obtain the state factor structure diagram.

6. The adaptive real-time optimized vector steering wheel cooperative control method according to claim 4 is characterized in that: The state influence evaluation of the real-time factor parameter set is performed in combination with the state factor structure diagram to obtain a real-time influence coefficient, including: Analyze and obtain the diagonal matrix of the state factor structure diagram; Analyze and obtain the adjacency matrix of the state factor structure graph; Calculating the Laplace matrix of the state factor structure graph according to the diagonal matrix and the adjacency matrix; Performing eigendecomposition processing on the Laplace matrix in combination with the real-time factor parameter set to obtain a state factor structure eigenvalue; The state factor structure characteristic value is recorded as the real-time influence coefficient.

7. The method for adaptive real-time optimized vector steering wheel cooperative control according to claim 1, characterized in that: include: Extracting a first control record of a first steering wheel module in the simulation control record, wherein the first steering wheel module refers to any one steering wheel module among the multiple steering wheel modules; Analyzing the first control record to obtain a first control energy consumption, and calculating a comprehensive control energy consumption based on the first control energy consumption; Comparing and analyzing the simulated actual trajectory in the simulated control record with the simulated predetermined trajectory to obtain the simulated control accuracy; Performing weighted calculation on the normalized comprehensive control energy consumption and the simulation control accuracy to obtain the simulation control fitness; When the simulation control fitness reaches a predetermined fitness threshold in the predetermined control constraint, the simulation control record satisfies the predetermined control constraint.

8. An adaptive real-time optimized vector steering wheel cooperative control system, characterized in that: The method for implementing the adaptive real-time optimized vector steering wheel cooperative control method according to any one of claims 1 to 7 comprises a plurality of steering wheel modules, and the system further comprises: Initial state information acquisition component: acquires initial state information of the vector steering wheel, wherein the initial state information includes steering wheel structure information and steering wheel load information; Initial state information analysis component: analyzing the steering wheel structure information and the steering wheel load information to obtain the initial dynamic state of the vector steering wheel; Dynamic monitoring component: reads the dynamic state factor set, and activates the state observer to dynamically monitor the dynamic state factor set of the vector steering wheel to obtain a real-time factor parameter set, wherein the dynamic state factor set includes multiple parameters related to the environment and operating conditions; A state prediction component: using the real-time factor parameter set and the initial dynamic state as input information of a state prediction model, and obtaining output information through the state prediction model, wherein the output information includes the predicted dynamic state of the vector steering wheel; A simulation control component: generating a coordinated control strategy for the plurality of steering wheel modules, and performing simulation control on the vector steering wheel under the predicted dynamic state based on the coordinated control strategy to obtain a simulation control record; Optimal strategy generation component: when the simulated control record satisfies the predetermined control constraint, the collaborative control strategy is used as the optimal strategy; Adaptive real-time control component: adaptively controls the vector steering wheel in real time according to the optimal strategy.

Images