Omnidirectional mobile robot driving control method based on non-linear adaptive control

Abstract

The invention discloses an omnidirectional mobile robot driving control method based on nonlinear adaptive control, which comprises the following steps: acquiring a differential signal of an expected trajectory signal input by an omnidirectional mobile robot, and determining an expected speed input quantity; acquiring a driving signal of the omnidirectional mobile robot; estimating the wheel slip interference of the omnidirectional mobile robot in real time according to the expected speed input quantity and the driving signal to obtain estimation information; judging whether the wheel slips according to the estimation information; when it is determined that the wheels do not slip, a kinematics control law is designed through a backstepping method; and in combination with a kinematics control law and active disturbance rejection nonlinearity, obtaining control input of the omni-directional mobile robot, and carrying out driving control on the omni-directional mobile robot. And the omni-directional mobile robot is accurately controlled to stably run.

Description

Technical Field

[0001] The invention relates to the technical field of robots, and in particular to a driving control method of an omnidirectional mobile robot based on nonlinear adaptive control. Background Art

[0002] In actual working environments, omnidirectional mobile robots often slip due to reasons such as slippery ground, ice, and aging of the outer rubber of the wheels, which seriously affects the position and motion state of the robot. The omnidirectional mobile robot is a multi-input and multi-output nonlinear system. During the trajectory tracking control process, the robot's position and motion state are affected by wheel slippage. The self-disturbance rejection controller has many parameters and is coupled. The robustness of the controller is often affected to a certain extent due to improper parameter adjustment. The existing technology cannot accurately control the stable driving of the omnidirectional mobile robot. Summary of the invention

[0003] The present invention aims to solve at least one of the technical problems in the above-mentioned technology to a certain extent. To this end, the purpose of the present invention is to propose a driving control method of an omnidirectional mobile robot based on nonlinear adaptive control to accurately control the omnidirectional mobile robot to drive stably.

[0004] To achieve the above object, an embodiment of the present invention proposes a driving control method of an omnidirectional mobile robot based on nonlinear adaptive control, comprising:

[0005] Obtaining a differential signal of a desired trajectory signal input by the omnidirectional mobile robot to determine a desired speed input;

[0006] Obtaining the driving signal of the omnidirectional mobile robot;

[0007] The wheel slip disturbance of the omnidirectional mobile robot is estimated in real time according to the expected speed input and the driving signal to obtain estimation information;

[0008] Determine whether the wheel is slipping based on the estimated information;

[0009] When it is determined that the wheels are not slipping, the kinematic control law is designed using the backstepping method;

[0010] The kinematic control law and the anti-disturbance nonlinearity are combined to obtain the control input of the omnidirectional mobile robot, and the driving of the omnidirectional mobile robot is controlled.

[0011] According to some embodiments of the present invention, a nonlinear tracking differentiator based on a sliding mode technique is used to arrange a transition process and obtain a differential signal of a desired trajectory signal input by the omnidirectional mobile robot.

[0012] According to some embodiments of the present invention, a design method of a nonlinear tracking differentiator based on sliding mode technology includes:

[0013] According to the system characteristics and control objectives of the omnidirectional mobile robot, a sliding surface is designed so that the system state can slide along the sliding surface.

[0014] Based on the design of sliding surface, a sliding mode control law is constructed to drive the system state to move toward the sliding surface and keep it stable on the sliding surface.

[0015] The sliding mode control law is combined with a nonlinear tracking differentiator, and the robustness of the sliding mode control law is used to suppress the influence of external disturbances and parameter changes on the extraction of differential signals.

[0016] According to some embodiments of the present invention, obtaining a differential signal of a desired trajectory signal input by an omnidirectional mobile robot and determining a desired speed input amount includes:

[0017] Determine the desired trajectory signal according to the position information and orientation information of the omnidirectional mobile robot in two-dimensional space;

[0018] The derivative of the desired trajectory signal is calculated to determine the linear velocity and angular velocity of the omnidirectional mobile robot as the desired velocity input.

[0019] According to some embodiments of the present invention, the wheel slip disturbance of the omnidirectional mobile robot is estimated in real time according to the expected speed input and the driving signal to obtain the estimation information, including:

[0020] Constructing an extended state observer;

[0021] The extended state observer receives the desired speed input as an output signal; the extended state observer receives the driving signal of the omnidirectional mobile robot as an input signal; based on the extended state observer, the disturbance to the system of the omnidirectional mobile robot is expanded into a system state, and the state variables and the expanded state of the system are estimated through an internal dynamic update process according to the output signal and the input signal, and the wheel slip interference is estimated in real time to obtain the estimated information.

[0022] According to some embodiments of the present invention, constructing an extended state observer includes:

[0023] Model the wheel slip disturbance as a disturbance term of the system and analyze the disturbance types in the system, including internal dynamics and external disturbances;

[0024] The uncertainties or disturbances in the system are regarded as new state variables, and an augmented state space is constructed. Based on the augmented state space, a new state equation is established; the state equation will contain the state variables of the original system and the newly added augmented state variables;

[0025] Select a nonlinear observer as the state observer according to system characteristics and control requirements;

[0026] According to the state equation, the observation equation of the extended state observer is constructed, and the observation equation will be used to estimate the state and extended state of the system according to the output and control input of the system;

[0027] The gain matrix of the state observer is designed to ensure that the observation error can converge to zero quickly over time. The parameters of the state observer are tuned to obtain the extended state observer.

[0028] According to some embodiments of the present invention, parameter tuning of a state observer includes:

[0029] Initial parameter setting: According to the system model and design experience, the initial parameters of the state observer are set, including the observer bandwidth and the expanded state gain;

[0030] Parameter optimization: adjust the state observer parameters through simulation or experiment to optimize the state observer performance, including improving the convergence speed, reducing the steady-state error and enhancing the anti-disturbance capability;

[0031] Online adjustment: Adjust the state observer parameters in real time according to the system state and observation error.

[0032] According to some embodiments of the present invention, when it is determined that the wheel does not slip, a kinematic control law is designed using a backstepping method, including:

[0033] Construct a kinematic model of an omnidirectional mobile robot;

[0034] Determine a control target of the omnidirectional mobile robot according to a kinematic model; the control target is that the omnidirectional mobile robot moves at a predetermined position and speed;

[0035] The kinematic model is decomposed into multiple subsystems, each of which is a subset of the system state variables. For each subsystem, a virtual control quantity is designed, starting from the output error of the system, and by introducing virtual control input, the design is gradually reversed until the actual control input is reached. By selecting appropriate virtual control input to ensure that the derivative of the Lyapunov function is negative, the deduction is gradually backwards until the real control input is obtained, and the kinematic control law is obtained based on the derivation results of the backstepping method.

[0036] According to some embodiments of the present invention, the further comprising:

[0037] When it is determined that the wheel is slipping, the vibration signal of the drive motor corresponding to each wheel is collected;

[0038] Calculate the energy entropy of the drive motor corresponding to each wheel according to the vibration signal of the drive motor corresponding to each wheel, and compare it with the preset energy entropy threshold;

[0039] The wheels whose energy entropy is greater than the preset energy entropy threshold are marked as slipping wheels and displayed.

[0040] According to some embodiments of the present invention, calculating the energy entropy of the drive motor corresponding to each wheel according to the vibration signal of the drive motor corresponding to each wheel includes:

[0041]

[0042] Where H is the energy entropy of the drive motor; w i is the energy loss value of the i-th vibration signal of the driving motor collected; is the average energy loss value of the n vibration signals collected from the drive motor; q i is the energy loss value of the i-th vibration signal after EEMD processing; is the average energy loss value of n vibration signals after EEMD processing; n is the number of vibration signals of the drive motor collected; E i is the energy value of the i-th vibration signal of the driving motor collected; E is the total energy value of the n vibration signals of the driving motor collected.

[0043] The present invention proposes an omnidirectional mobile robot driving control method based on nonlinear adaptive control. The wheel slip interference of the omnidirectional mobile robot is estimated in real time according to the expected speed input and the driving signal to obtain the estimation information. When it is determined that the wheels do not slip, the kinematic control law is designed by the backstepping method. The kinematic control law is combined with the self-disturbance rejection nonlinearity to obtain the control input of the omnidirectional mobile robot, and the driving control of the omnidirectional mobile robot is performed, thereby improving the accuracy of the driving control of the omnidirectional mobile robot.

[0044] Other features and advantages of the present invention will be described in the following description, and partly become apparent from the description, or understood by practicing the present invention. The purpose and other advantages of the present invention can be realized and obtained by the structures particularly pointed out in the written description and the accompanying drawings.

[0045] The technical solution of the present invention is further described in detail below through the accompanying drawings and embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] The accompanying drawings are used to provide a further understanding of the present invention and constitute a part of the specification. Together with the embodiments of the present invention, they are used to explain the present invention and do not constitute a limitation of the present invention. In the accompanying drawings:

[0047] Figure 1 is a flow chart of a driving control method of an omnidirectional mobile robot based on nonlinear adaptive control according to an embodiment of the present invention;

[0048] Figure 2 is a flow chart for determining a desired speed input according to one embodiment of the present invention;

[0049] Figure 3 is a flow chart of obtaining estimation information according to an embodiment of the present invention. DETAILED DESCRIPTION

[0050] The preferred embodiments of the present invention are described below in conjunction with the accompanying drawings. It should be understood that the preferred embodiments described herein are only used to illustrate and explain the present invention, and are not used to limit the present invention.

[0051] like Figure 1 As shown, the embodiment of the present invention proposes a driving control method for an omnidirectional mobile robot based on nonlinear adaptive control, comprising steps S1-S6:

[0052] S1, obtaining a differential signal of the desired trajectory signal input by the omnidirectional mobile robot, and determining the desired speed input;

[0053] S2, obtaining a driving signal of the omnidirectional mobile robot;

[0054] S3, performing real-time estimation of the wheel slip interference of the omnidirectional mobile robot according to the expected speed input and the driving signal to obtain estimation information;

[0055] S4, judging whether the wheel is slipping according to the estimated information;

[0056] S5. When it is determined that the wheels are not slipping, the kinematic control law is designed using the backstepping method;

[0057] S6. Combining the kinematic control law with the anti-disturbance nonlinearity, the control input of the omnidirectional mobile robot is obtained, and the driving of the omnidirectional mobile robot is controlled.

[0058] The working principle of the above technical solution: In this embodiment, the desired speed input is an output signal; the drive signal is generated by a drive motor included in the omnidirectional mobile robot. The drive signal of the omnidirectional mobile robot is obtained based on the motor drive circuit, and the motor drive circuit uses PWM (pulse width modulation) technology to control the speed and direction of the motor. The controller controls the output of the motor driver by outputting PWM signals with different duty cycles, thereby achieving precise control of the motor speed.

[0059] In this embodiment, the wheel slip disturbance of the omnidirectional mobile robot is estimated in real time based on the expected speed input and the drive signal. In this process, a nonlinear tracking differentiator based on the sliding mode technology is used to arrange the transition process to improve the accuracy of signal processing and obtain estimation information.

[0060] In this embodiment, judging whether the wheel is slipping based on the estimated information includes: speed comparison: calculating the difference between the actual speed measured by the encoder and the expected speed issued by the controller. If the actual speed is much lower than the expected speed, slipping may have occurred. Position comparison: integrating the speed measured by the encoder over a period of time to obtain the actual position of the wheel and compare it with the expected position. If there is a significant deviation between the actual position and the expected position, slipping may have occurred.

[0061] In this embodiment, in order to improve the robustness of the closed-loop control system when the wheels of the omnidirectional mobile robot slip, the backstepping method is used to design the kinematic control law when the wheels do not slip. The backstepping method is a technique that decomposes a complex nonlinear system into multiple smaller subsystems and designs intermediate virtual control quantities for each subsystem. These virtual control quantities are then "back-pushed" to the entire system to design the overall control law of the system. The control input of the omnidirectional mobile robot is obtained by combining the kinematic control law with the self-disturbance rejection nonlinearity, which improves the accuracy of determining the control input, thereby improving the accuracy of driving control of the omnidirectional mobile robot.

[0062] The beneficial effects of the above technical solution are as follows: the wheel slip interference of the omnidirectional mobile robot is estimated in real time according to the expected speed input and the driving signal to obtain the estimation information; when it is determined that the wheels are not slipping, the kinematic control law is designed by the backstepping method; the kinematic control law is combined with the self-disturbance rejection nonlinearity to obtain the control input of the omnidirectional mobile robot, and the driving control of the omnidirectional mobile robot is performed, thereby improving the accuracy of the driving control of the omnidirectional mobile robot.

[0063] According to some embodiments of the present invention, a nonlinear tracking differentiator based on a sliding mode technique is used to arrange a transition process and obtain a differential signal of a desired trajectory signal input by the omnidirectional mobile robot.

[0064] Working principle and beneficial effects of the above technical solution: The nonlinear tracking differentiator (NTD) based on sliding mode technology plays an important role in control systems, especially in situations where a transition process needs to be arranged to smooth the reference signal and its differential signal. The main function of the nonlinear tracking differentiator (NTD) is to arrange a transition process for the input signal (such as the desired trajectory or reference signal) and extract the differential signal of the signal at the same time. This transition process helps to reduce the system's sensitivity to changes in the input signal and improve the stability and robustness of the system. Specifically, NTD can: Smooth the input signal: By introducing a transition process, the change of the input signal is made smoother and the impact of sudden changes on the system is reduced. Extract the differential signal: During the transition process, NTD can accurately and stably extract the differential signal of the input signal, which is particularly important for systems that require differential signals as control inputs. Sliding mode technology is a variable structure control method that makes the system state reach the sliding surface within a finite time and keep sliding on the sliding surface by designing appropriate sliding surfaces and control laws. The nonlinear tracking differentiator based on sliding mode technology has the following characteristics: Switching function: The switching function is introduced in NTD to realize the rapid tracking of the input signal and the extraction of the differential signal. The existence of the switching function enables the system state to be adjusted quickly when it approaches the sliding surface, thereby reducing the tracking error. Dynamic adjustment: The parameters of NTD (such as sliding mode gain, parameters of the switching function, etc.) can be dynamically adjusted according to the actual operating state of the system to adapt to different control requirements and environmental changes. Robustness: Due to the characteristics of sliding mode technology, NTD has strong robustness to changes in system parameters and external disturbances. Even if there are uncertainties and interferences in the system, NTD can still maintain good tracking performance and differential accuracy. In the control of omnidirectional mobile robots, the nonlinear tracking differentiator based on sliding mode technology can be used to arrange the transition process of the robot's desired trajectory signal and extract its differential signal as a speed control instruction. The specific steps are as follows: Input the desired trajectory: Input the desired trajectory signal (such as position, speed, etc.) into NTD. Arrange the transition process: NTD automatically arranges a transition process according to the input signal to make the change of the desired trajectory smoother. Extract differential signals: During the transition process, NTD simultaneously extracts the differential signals of the desired trajectory (such as speed, acceleration, etc.). Output control instructions: The extracted differential signals are input into the robot controller as control instructions to achieve accurate trajectory tracking and speed control. Based on the nonlinear tracking differentiator of sliding mode technology, NTD can effectively improve the stability and robustness of the system and meet the requirements of complex control tasks by arranging the transition process for the input signal and extracting the differential signal. In the control of omnidirectional mobile robots, the robot's trajectory tracking accuracy and dynamic response speed are significantly improved.

[0065] According to some embodiments of the present invention, a design method of a nonlinear tracking differentiator based on sliding mode technology includes:

[0066] According to the system characteristics and control objectives of the omnidirectional mobile robot, a sliding surface is designed so that the system state can slide along the sliding surface.

[0067] Based on the design of sliding surface, a sliding mode control law is constructed to drive the system state to move toward the sliding surface and keep it stable on the sliding surface.

[0068] The sliding mode control law is combined with a nonlinear tracking differentiator, and the robustness of the sliding mode control law is used to suppress the influence of external disturbances and parameter changes on the extraction of differential signals.

[0069] The working principle and beneficial effects of the above technical solution are as follows: A sliding surface based on position and velocity errors is designed to ensure that the system state can slide along this surface to the desired trajectory. The position error is defined as e x =x q -x and e y =y q -y; where x q With y q is the desired position; (x, y) is the position of the omnidirectional mobile robot, (V x , V y ) is the speed of the omnidirectional mobile robot. The speed error is defined as: and e x 'with e y ' is the desired speed; the sliding surface is:

[0070] S=c1e x +c2e x '+c3e y +c4e y ', c1, c2, c3, c4 are positive constants that need to be adjusted according to the system characteristics and control objectives to ensure the stability and convergence of the sliding surface. The selection of these parameters is usually based on the dynamic characteristics of the system and the requirements for convergence speed.

[0071] The purpose of the sliding mode control law is to drive the system state to move toward the sliding surface and keep it stable on the sliding surface. This is achieved by introducing a discontinuous control term (i.e., switching term), U=U1+U2; U1 is an equivalent control term used to keep the system state on the sliding surface; U2 is a switching control term used to pull the system state toward the sliding surface. Equivalent control term: obtained by setting the sliding surface S to zero and solving the control input U, which involves the inverse operation of the system dynamic equation.

[0072] Switching control term: designed as a function related to the sign of the sliding surface S to ensure that when S = 0, the control input can quickly pull the system state back to the sliding surface, in the form of a sign function sgn(s) or its continuous approximation (such as a saturation function). For example, U2 = -ksgn(s), k is a positive constant used to adjust the strength of the switching control.

[0073] Combining the nonlinear tracking differentiator (NTD) with the sliding mode control law can take advantage of the smooth transition and differential signal extraction capabilities of the NTD, while using the robustness of the sliding mode control to suppress external disturbances and parameter changes. Combination method: Input processing: The desired trajectory signal (x q ,y q ) is input into NTD to obtain a smooth transition trajectory and the corresponding differential signal Control law design: Use the smooth trajectory and derivative signal of the NTD output as the desired signal in the sliding mode control law. In this way, the sliding mode control law will generate control inputs based on these smoother and more accurate signals. Robustness enhancement: Due to the robustness of the sliding mode control law, even if there are external disturbances or system parameter changes, the system state can still remain on the sliding surface or be quickly pulled back to the sliding surface, thereby ensuring the stability and performance of the system.

[0074] Through this combination, the omnidirectional mobile robot can track the desired trajectory more accurately and effectively suppress the impact of external interference and parameter changes on system performance.

[0075] like Figure 2 As shown, according to some embodiments of the present invention, obtaining a differential signal of a desired trajectory signal input by an omnidirectional mobile robot and determining a desired speed input amount include steps S11-S12:

[0076] S11, determining a desired trajectory signal according to the position information and orientation information of the omnidirectional mobile robot in the two-dimensional space;

[0077] S12. Calculate the derivative of the desired trajectory signal to determine the linear velocity and angular velocity of the omnidirectional mobile robot as the desired velocity input.

[0078] The beneficial effects of the above technical solution are as follows: the position information includes (x, y), and the orientation information is θ; the information of the expected trajectory signal changing with time is calculated, the derivative of the expected trajectory signal is calculated, and the linear velocity and angular velocity of the omnidirectional mobile robot are determined as the expected velocity input. It is convenient to accurately determine the expected velocity input.

[0079] like Figure 3As shown, according to some embodiments of the present invention, the wheel slip disturbance of the omnidirectional mobile robot is estimated in real time according to the expected speed input and the driving signal to obtain the estimation information, including steps S31-S32:

[0080] S31, construct an extended state observer;

[0081] S32, the expanded state observer receives the desired speed input as an output signal; the expanded state observer receives the driving signal of the omnidirectional mobile robot as an input signal; based on the expanded state observer, the disturbance to the system of the omnidirectional mobile robot is expanded into a system state, and the state variables and the expanded state of the system are estimated through an internal dynamic update process according to the output signal and the input signal, and the wheel slip interference is estimated in real time to obtain estimation information.

[0082] The working principle and beneficial effects of the above technical solution: The extended state observer is a dynamic process. It does not rely on the precise mathematical model of the control system. It only needs the input information and output information of the control object to estimate the state variables of the system and the unknown dynamic disturbance of the system in real time. According to the output signal and the input signal, through the internal dynamic update process, the state variables and extended state of the system are estimated, and the wheel slip interference is estimated in real time to obtain the estimated information. Improve the accuracy of the estimated information.

[0083] According to some embodiments of the present invention, constructing an extended state observer includes:

[0084] Model the wheel slip disturbance as a disturbance term of the system and analyze the disturbance types in the system, including internal dynamics and external disturbances;

[0085] The uncertainties or disturbances in the system are regarded as new state variables, and an augmented state space is constructed. Based on the augmented state space, a new state equation is established; the state equation will contain the state variables of the original system and the newly added augmented state variables;

[0086] Select a nonlinear observer as the state observer according to system characteristics and control requirements;

[0087] According to the state equation, the observation equation of the extended state observer is constructed, and the observation equation will be used to estimate the state and extended state of the system according to the output and control input of the system;

[0088] The gain matrix of the state observer is designed to ensure that the observation error can converge to zero quickly over time. The parameters of the state observer are tuned to obtain the extended state observer.

[0089] The working principle and beneficial effects of the above technical solution: External disturbances: mainly include disturbances caused by external environmental factors such as wheel slippage, uneven ground, wind resistance, etc. Internal disturbances: mainly caused by uncertainty and dynamic characteristics within the system, such as nonlinearity of the motor model, sensor noise, friction of mechanical parts, etc. In order to deal with these disturbances, we can regard them as new state variables and construct an augmented state space. In this space, the state vector of the system will include the state variables of the original system (such as position, speed) and the newly added expanded state variables (representing disturbances).

[0090] Augmented state vector: Let the state vector of the original system be x = [x1, x2, …, x n ] T (where x1, x2, …, x n It may be position, velocity, etc.), the newly added extended state variable is w (representing disturbance), then the augmented state vector is: e =[x1,x2,…,x n ,w] T ;

[0091] New state equation: Based on the augmented state vector, a new state equation can be established:

[0092] x˙ e =A e x e +B e u+E e d, where A e and B e is the new system matrix, d is any other possible external disturbance (in addition to the disturbance modeled as state w), E e is the transfer matrix of these disturbances to the system state. In practical applications, if the effect of d is relatively small or can be ignored, d and E can be excluded. e .

[0093] Select a nonlinear observer;

[0094] Considering the possible nonlinearity and uncertainty in the system, it is appropriate to choose a nonlinear observer (such as a nonlinear extended state observer) as the state observer. The nonlinear observer can better handle the nonlinear characteristics in the system and provide more accurate state estimation.

[0095] The observation equations estimate the state and expansion state of the system based on the system outputs and control inputs. For an omnidirectional mobile robot, the outputs may include sensor measurements such as position and velocity.

[0096] Observation equation:

[0097]

[0098] is an estimate of xe, L is the observer gain matrix, which is used to adjust the dynamic characteristics of the observer to ensure the accuracy of the estimate. u is the control input; Ce is the system matrix describing the state vector; is the estimated information of the state vector; y is the performance information of the state vector.

[0099] Designing the observer gain matrix L is the key to ensuring fast convergence of the observation error. It is a trade-off between an in-depth understanding of the system dynamics and the observer performance indicators. The gain matrix L can be determined by pole placement, linear matrix inequality (LMI) optimization, or empirical trial and error. Parameter tuning: After determining the preliminary value of the gain matrix L, the observer needs to be parameterized through simulation or experiment to optimize its performance. This includes adjusting the elements of the gain matrix to balance the observer's convergence speed, stability, and robustness to noise.

[0100] Through the above steps, an extended state observer for an omnidirectional mobile robot is constructed, which can estimate the state and extended state (i.e., disturbance) of the system in real time, providing accurate information for subsequent controller design.

[0101] According to some embodiments of the present invention, parameter tuning of a state observer includes:

[0102] Initial parameter setting: According to the system model and design experience, the initial parameters of the state observer are set, including the observer bandwidth and the expanded state gain;

[0103] Parameter optimization: adjust the state observer parameters through simulation or experiment to optimize the state observer performance, including improving the convergence speed, reducing the steady-state error and enhancing the anti-disturbance capability;

[0104] Online adjustment: Adjust the state observer parameters in real time according to the system state and observation error.

[0105] Working principle and beneficial effects of the above technical solution: Parameter tuning is an iterative process that aims to optimize the convergence speed, steady-state error and anti-disturbance ability of the observer by adjusting its parameters. Initial parameter setting, system model analysis: First, it is necessary to have a deep understanding of the dynamic characteristics of the system, including its stability, response speed and possible nonlinearity or uncertainty. Design experience: Based on previous design experience or parameter settings of similar systems, the parameters of the state observer can be preliminarily set. Especially for the extended state observer, special attention should be paid to the gain setting of the extended state (i.e., the disturbance term). Observer bandwidth: The observer bandwidth is an important parameter that determines the response speed of the observer to changes in the system state. Setting the bandwidth too high may make the observer sensitive to noise, while setting it too low may make the observer respond too slowly. Extended state gain: The gain of the extended state (i.e., the disturbance term) determines the observer's ability to estimate disturbances. This gain needs to be large enough to quickly capture disturbances, but not too large to avoid introducing unnecessary noise. Parameter optimization: Simulation test: Before applying the observer to the actual system, its performance is first evaluated through simulation tests. In the simulation environment, the parameters of the observer can be easily adjusted and their impact on the system state estimation can be observed. Performance indicators: A series of performance indicators are defined to evaluate the performance of the observer, such as convergence speed, steady-state error, overshoot, anti-disturbance ability, etc. These indicators are consistent with the control objectives and performance requirements of the system. Parameter adjustment: According to the simulation results and performance indicators, the parameters of the observer are gradually adjusted. Usually, this is a trial-and-error process, and multiple iterations are required to find the optimal parameter combination. Optimization algorithm: If the system is complex or the parameter adjustment range is large, an optimization algorithm (such as genetic algorithm, particle swarm optimization, etc.) is used to automatically search for the optimal parameters. Online adjustment: Real-time feedback: In actual applications, the system state and observation error may change with time and environmental conditions. Therefore, the parameters of the observer also need to be able to be adjusted in real time according to the actual situation. Adaptive mechanism: An adaptive mechanism is introduced to adjust the parameters of the observer online. This can be achieved by monitoring signals such as observation error, system state or external disturbance, and dynamically adjusting the parameters of the observer according to the changes in these signals. Safety restrictions: When adjusting the observer parameters, the stability and safety of the system need to be ensured. Therefore, it is necessary to set a reasonable parameter adjustment range and restriction conditions to avoid system instability or damage due to improper parameter adjustment. Through reasonable initial parameter setting, parameter optimization and online adjustment, a state observer with excellent performance and strong robustness can be constructed.

[0106] According to some embodiments of the present invention, when it is determined that the wheel does not slip, a kinematic control law is designed using a backstepping method, including:

[0107] Construct a kinematic model of an omnidirectional mobile robot;

[0108] Determine a control target of the omnidirectional mobile robot according to a kinematic model; the control target is that the omnidirectional mobile robot moves at a predetermined position and speed;

[0109] The kinematic model is decomposed into multiple subsystems, each of which is a subset of the system state variables. For each subsystem, a virtual control quantity is designed, starting from the output error of the system, and by introducing virtual control input, the design is gradually reversed until the actual control input is reached. By selecting appropriate virtual control input to ensure that the derivative of the Lyapunov function is negative, the deduction is gradually backwards until the real control input is obtained, and the kinematic control law is obtained based on the derivation results of the backstepping method.

[0110] The working principle and beneficial effects of the above technical solution: Starting from the last subsystem, the designed virtual control quantity is gradually back-stepped to the previous subsystem until the overall control law of the entire system is finally obtained. In this process, the virtual control quantity of each subsystem will become the "input" of the next subsystem. The subsystems include position subsystems, direction subsystems, and speed subsystems. Then, a virtual control quantity is designed for each subsystem, and the overall control laws ω and v are obtained by back-stepping. Appropriate control gains need to be selected to ensure that the derivative of the Lyapunov function is negative, thereby ensuring the stability and convergence of the system. The key to the backstepping method is to "backward" design the controller by gradually introducing virtual control quantities until the actual physical control input is reached.

[0111] According to some embodiments of the present invention, the further comprising:

[0112] When it is determined that the wheel is slipping, the vibration signal of the drive motor corresponding to each wheel is collected;

[0113] Calculate the energy entropy of the drive motor corresponding to each wheel according to the vibration signal of the drive motor corresponding to each wheel, and compare it with the preset energy entropy threshold;

[0114] The wheels whose energy entropy is greater than the preset energy entropy threshold are marked as slipping wheels and displayed.

[0115] The working principle and beneficial effects of the above technical solution are as follows: when determining that the wheel is slipping, the vibration signal of the drive motor corresponding to each wheel is collected; based on the vibration signal of the drive motor corresponding to each wheel, the energy entropy of the drive motor corresponding to each wheel is calculated, and compared with the preset energy entropy threshold; the wheels whose energy entropy is greater than the preset energy entropy threshold are considered to have low energy conversion efficiency and slippage, and are marked as slipping wheels and displayed, so that the user can clearly understand which wheel is slipping at a certain time and position, and at the same time set corresponding control strategies for the slipping wheels to improve driving stability.

[0116] According to some embodiments of the present invention, calculating the energy entropy of the drive motor corresponding to each wheel according to the vibration signal of the drive motor corresponding to each wheel includes:

[0117]

[0118] Where H is the energy entropy of the drive motor; w i is the energy loss value of the i-th vibration signal of the driving motor collected; is the average energy loss value of the n vibration signals collected from the drive motor; q i is the energy loss value of the i-th vibration signal after EEMD processing; q is the average energy loss value of n vibration signals after EEMD processing; n is the number of vibration signals of the drive motor collected; E i is the energy value of the i-th vibration signal of the driving motor collected; E is the total energy value of the n vibration signals of the driving motor collected.

[0119] The working principle and beneficial effects of the above technical solution: the energy entropy of the drive motor is accurately calculated based on parameters such as the energy loss value of the collected vibration signal of the drive motor, the average energy loss value of the collected vibration signals of the drive motor, and the number of collected vibration signals of the drive motor, thereby improving the accuracy of judging the energy entropy and the preset energy entropy threshold, and further improving the accuracy of judging which is the slipping wheel.

[0120] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include these modifications and variations.

Claims

1. A driving control method for an omnidirectional mobile robot based on nonlinear adaptive control, characterized in that: include: Obtaining a differential signal of a desired trajectory signal input by the omnidirectional mobile robot to determine a desired speed input; Obtaining the driving signal of the omnidirectional mobile robot; The wheel slip disturbance of the omnidirectional mobile robot is estimated in real time according to the expected speed input and the driving signal to obtain estimation information; Determine whether the wheel is slipping based on the estimated information; When it is determined that the wheels are not slipping, the kinematic control law is designed using the backstepping method; The kinematic control law and the anti-disturbance nonlinearity are combined to obtain the control input of the omnidirectional mobile robot, and the driving of the omnidirectional mobile robot is controlled.

2. The driving control method of an omnidirectional mobile robot based on nonlinear adaptive control as claimed in claim 1, characterized in that: A nonlinear tracking differentiator based on sliding mode technology is used to arrange the transition process and obtain the differential signal of the desired trajectory signal input by the omnidirectional mobile robot.

3. The driving control method of an omnidirectional mobile robot based on nonlinear adaptive control as claimed in claim 2, characterized in that: The design method of nonlinear tracking differentiator based on sliding mode technology includes: According to the system characteristics and control objectives of the omnidirectional mobile robot, a sliding surface is designed so that the system state can slide along the sliding surface. Based on the design of sliding surface, a sliding mode control law is constructed to drive the system state to move toward the sliding surface and keep it stable on the sliding surface. The sliding mode control law is combined with a nonlinear tracking differentiator, and the robustness of the sliding mode control law is used to suppress the influence of external disturbances and parameter changes on the extraction of differential signals.

4. The driving control method of an omnidirectional mobile robot based on nonlinear adaptive control as claimed in claim 1, characterized in that: Obtain the differential signal of the desired trajectory signal input by the omnidirectional mobile robot and determine the desired speed input, including: Determine the desired trajectory signal according to the position information and orientation information of the omnidirectional mobile robot in two-dimensional space; The derivative of the desired trajectory signal is calculated to determine the linear velocity and angular velocity of the omnidirectional mobile robot as the desired velocity input.

5. The driving control method of an omnidirectional mobile robot based on nonlinear adaptive control as claimed in claim 1, characterized in that: The wheel slip disturbance of the omnidirectional mobile robot is estimated in real time according to the expected speed input and the driving signal to obtain the estimated information, including: Constructing an extended state observer; The extended state observer receives the desired speed input as an output signal; the extended state observer receives the driving signal of the omnidirectional mobile robot as an input signal; based on the extended state observer, the disturbance to the system of the omnidirectional mobile robot is expanded into a system state, and the state variables and the expanded state of the system are estimated through an internal dynamic update process according to the output signal and the input signal, and the wheel slip interference is estimated in real time to obtain the estimated information.

6. The driving control method of an omnidirectional mobile robot based on nonlinear adaptive control as claimed in claim 5, characterized in that: Construct an extended state observer, including: Model the wheel slip disturbance as a disturbance term of the system and analyze the disturbance types in the system, including internal dynamics and external disturbances; The uncertainties or disturbances in the system are regarded as new state variables, and an augmented state space is constructed. Based on the augmented state space, a new state equation is established; the state equation will contain the state variables of the original system and the newly added augmented state variables; Select a nonlinear observer as the state observer according to system characteristics and control requirements; According to the state equation, the observation equation of the extended state observer is constructed, and the observation equation will be used to estimate the state and extended state of the system according to the output and control input of the system; The gain matrix of the state observer is designed to ensure that the observation error can converge to zero quickly over time. The parameters of the state observer are tuned to obtain the extended state observer.

7. The driving control method of an omnidirectional mobile robot based on nonlinear adaptive control as claimed in claim 6, characterized in that: Parameter tuning of the state observer includes: Initial parameter setting: According to the system model and design experience, the initial parameters of the state observer are set, including the observer bandwidth and the expanded state gain; Parameter optimization: adjust the state observer parameters through simulation or experiment to optimize the state observer performance, including improving the convergence speed, reducing the steady-state error and enhancing the anti-disturbance capability; Online adjustment: Adjust the state observer parameters in real time according to the system state and observation error.

8. The driving control method of an omnidirectional mobile robot based on nonlinear adaptive control as claimed in claim 1, characterized in that: When the wheels are determined not to slip, the backstepping method is used to design the kinematic control law, including: Construct a kinematic model of an omnidirectional mobile robot; Determine a control target of the omnidirectional mobile robot according to a kinematic model; the control target is that the omnidirectional mobile robot moves at a predetermined position and speed; The kinematic model is decomposed into multiple subsystems, each of which is a subset of the system state variables. For each subsystem, a virtual control quantity is designed, starting from the output error of the system, and by introducing virtual control input, the design is gradually reversed until the actual control input is reached. By selecting appropriate virtual control input to ensure that the derivative of the Lyapunov function is negative, the deduction is gradually backwards until the real control input is obtained, and the kinematic control law is obtained based on the derivation results of the backstepping method.

9. The driving control method of an omnidirectional mobile robot based on nonlinear adaptive control as claimed in claim 1, characterized in that: Also includes: When it is determined that the wheel is slipping, the vibration signal of the drive motor corresponding to each wheel is collected; Calculate the energy entropy of the drive motor corresponding to each wheel according to the vibration signal of the drive motor corresponding to each wheel, and compare it with the preset energy entropy threshold; The wheels whose energy entropy is greater than the preset energy entropy threshold are marked as slipping wheels and displayed.

10. The driving control method of an omnidirectional mobile robot based on nonlinear adaptive control as claimed in claim 9, characterized in that: According to the vibration signal of the drive motor corresponding to each wheel, the energy entropy of the drive motor corresponding to each wheel is calculated, including: Where H is the energy entropy of the drive motor; w i is the energy loss value of the i-th vibration signal of the driving motor collected; is the average energy loss value of the n vibration signals collected from the drive motor; q i is the energy loss value of the i-th vibration signal after EEMD processing; is the average energy loss value of n vibration signals after EEMD processing; n is the number of vibration signals of the drive motor collected; E i is the energy value of the i-th vibration signal of the driving motor collected; E is the total energy value of the n vibration signals of the driving motor collected.

Images