Neural Network Sliding Mode Control Research Articles

Neural network sliding mode controller of atomic force microscope-based manipulation with different cantilever probes.

Development of nanotechnology has given rise to various applications, including the nano-manipulation process within small-size environments. The implementation of such processes requires the use of tools and proper equipment and understanding of various factors influencing it. One such tool is the atomic force microscope (AFM) and its probe, used for imaging surfaces and manipulation tools. The AFM probe is the most important element of the AFM with a key role in system function. The dynamic analysis and control of AFM are necessary to increase efficiency. In this paper, a model of AFM is reviewed and rewritten by considering various cantilever probes, including rectangular, V-shaped, and dagger. The AFM actuator was modeled and analyzed on uncertain conditions. The position of the stage was controlled to the desired position through the desired motion profiles. To overcome the problem of model nonlinearity, a neural network (NN) sliding mode controller was used to optimize the controller parameter and provide the desired output. The simulation of system was performed by the effective parameters, its control was implemented, and the results were analyzed. The simulation revealed that the modified sliding mode controller with learnable NN improved controller performance by decreasing the rise time and eliminating the overshot.

A Dual-Loop Control Approach of Active Magnetic Bearing System for Rotor Tracking Control

Rotor tracking control, which can be implemented by active magnetic bearing (AMB) system with high precision, can realize many functions, such as attitude control and special surface processing. However, large-motion rotor tracking control is difficult to implemented, due to AMB's highly nonlinear characteristics. In this paper, a dual-loop neural network sliding mode control (DL-NNSMC) system of AMB is proposed for rotor radial tracking control. The complete model of the AMB system is established and the dual-loop control system is designed. A circuit model that considers the rotor motion is established and the model-based inner loop of current control is established, conjointly for dealing with the influence of rotor motion on the current response and the unknown characteristics of the power amplifier. In the outer loop, a nonlinear electromagnetic force model is applied and a wavelet neural network sliding mode control algorithm is designed for accurate position control. Two cases of rotor trajectory tracking are simulated, and the simulation results demonstrate the validity of the proposed control system for large-motion rotor tracking control and its far superior control performance in terms of precision compared with common approaches based on sliding mode control (SMC).

Neural Network Identification and Sliding Mode Control for Hysteresis Nonlinear System with Backlash‐Like Model

A new neural network sliding mode control (NNSMC) is proposed for backlash‐like hysteresis nonlinear system in this paper. Firstly, only one neural network is designed to estimate the unknown system states and hysteresis section instead of multiscale neural network at former researches since that can save computation and simplify the controller design. Secondly, a new NNSMC is proposed for the hysteresis nonlinearity where it does not need tracking error transformation. Finally, the Lyapunov functions are adopted to guarantee the stabilities of the identification and control strategies semiglobally uniformly ultimately bounded (UUB). Two cases simulations are proved the effectiveness of the presented identification approach and the performance of the NNSMC.

高超声速飞行器的反步滑模神经网络控制系统

高超声速飞行器的反步滑模神经网络控制系统

Parallel robot with fuzzy neural network sliding mode control

With the advancement in research on parallel robots, control theory is increasingly applied in the field of robotics. Owing to its robustness, sliding mode variable structure control is extensively used in parallel robots. This article presents an adaptive sliding mode control method for nonlinear systems. A parallel robot control model with adaptive fuzzy sliding mode control was designed based on a fuzzy neural network control theory, and simulation results demonstrate its effectiveness of the method.

Fuzzy Neural Network Sliding Mode Control Based Chaos Synchronization for a Class of Fractional Order Chaotic System

Fuzzy Neural Network Sliding Mode Control Based Chaos Synchronization for a Class of Fractional Order Chaotic System

Study on active vibration isolation system using neural network sliding mode control

In this paper, an active vibration isolation system based on hybrid algorithm is presented in a wide frequency band. Initially, a nonlinear magnetostrictive actuator model is used to establish the appropriate parameters by experiments, which make the actuator using in vibration isolation system work in a better linear dynamic performance, then the sliding mode algorithm modified by neural network, a hybrid algorithm is proposed as the active control controller, and its stability is also analyzed by Lyapunov theory. Furthermore, a dynamic virtual prototype model of active vibration isolation is established to carry out the co-simulation with Adams and Matlab/Simulink, and the results show that under the difference excitations, the neural network sliding mode controller exhibits a good control performance, and the active vibration isolation can effectively improve the vibration isolation effect, reduce the force transmitted to the base and broaden the vibration isolation bandwidth.

Combined Neural Network and PD Adaptive Tracking Controller for Ship Steering System

In this paper, a combined RBF neural network sliding mode control and PD adaptive tracking controller is proposed for controlling the directional heading course of a ship. Due to the high nonlinearity and uncertainty of the ship dynamics as well as the effect of wave disturbances a performance evaluation and ship controller design is stay difficult task. The Neural network used for adaptively learn the uncertain dynamics bounds of the ship and their output used as part of the control law moreover the PD term is used to reduce the effect of the approximation error inherited in the RBF networks. The stability of the system with the combined control law guaranteed through Lyapunov analysis. Numeric simulation results confirm the proposed controller provide good system stability and convergence.

Dynamic boundary layer based neural network quasi-sliding mode control for soft touching down on asteroid

Pinpoint landing is a critical step in some asteroid exploring missions. This paper is concerned with the descent trajectory control for soft touching down on a small irregularly-shaped asteroid. A dynamic boundary layer based neural network quasi-sliding mode control law is proposed to track a desired descending path. The asteroid’s gravitational acceleration acting on the spacecraft is described by the polyhedron method. Considering the presence of input constraint and unmodeled acceleration, the dynamic equation of relative motion is presented first. The desired descending path is planned using cubic polynomial method, and a collision detection algorithm is designed. To perform trajectory tracking, a neural network sliding mode control law is given first, where the sliding mode control is used to ensure the convergence of system states. Two radial basis function neural networks (RBFNNs) are respectively used as an approximator for the unmodeled term and a compensator for the difference between the actual control input with magnitude constraint and nominal control. To improve the chattering induced by the traditional sliding mode control and guarantee the reachability of the system, a specific saturation function with dynamic boundary layer is proposed to replace the sign function in the preceding control law. Through the Lyapunov approach, the reachability condition of the control system is given. The improved control law can guarantee the system state move within a gradually shrinking quasi-sliding mode band. Numerical simulation results demonstrate the effectiveness of the proposed control strategy.

Adaptive Neural Network Sliding Mode Control for Quad Tilt Rotor Aircraft

A novel neural network sliding mode control based on multicommunity bidirectional drive collaborative search algorithm (M-CBDCS) is proposed to design a flight controller for performing the attitude tracking control of a quad tilt rotors aircraft (QTRA). Firstly, the attitude dynamic model of the QTRA concerning propeller tension, channel arm, and moment of inertia is formulated, and the equivalent sliding mode control law is stated. Secondly, an adaptive control algorithm is presented to eliminate the approximation error, where a radial basis function (RBF) neural network is used to online regulate the equivalent sliding mode control law, and the novel M-CBDCS algorithm is developed to uniformly update the unknown neural network weights and essential model parameters adaptively. The nonlinear approximation error is obtained and serves as a novel leakage term in the adaptations to guarantee the sliding surface convergence and eliminate the chattering phenomenon, which benefit the overall attitude control performance for QTRA. Finally, the appropriate comparisons among the novel adaptive neural network sliding mode control, the classical neural network sliding mode control, and the dynamic inverse PID control are examined, and comparative simulations are included to verify the efficacy of the proposed control method.

Neural network sliding mode control with time-varying delay for a variable speed wind turbine

This paper presents a real time neural network sliding mode control with time-varying delay (RNNSM) design for a variable speed wind turbine. The sliding mode control (SMC) method can be used for nonlinear uncertain systems. However, it provides some drawbacks linked to chattering, due to the higher needed switching gain in the case of large uncertainties. In order to reduce this gain, artificial neural network (ANN) is used for the estimation of model unknown parts and hence enable a lower switching gain to be used. The ANN is turned online; therefore, we cannot implement this control law in real time because of the time-varying delay generated by the ANN in online learning. This work is a proposal to compensate this delay so as to ensure the control in real time and maintain the advantage of online learning; we propose to convert the original system to a system with time-varying input delay which will practically represent the online learning time. We use Lyapunov-Krasovskii functional to prove the stability of the closed loop systems, and the effectiveness of the designed method is illustrated in simulations by the comparison with integral sliding mode control (ISMC).

Neural network sliding mode control with time-varying delay for a variable speed wind turbine

This paper presents a real time neural network sliding mode control with time-varying delay (RNNSM) design for a variable speed wind turbine. The sliding mode control (SMC) method can be used for nonlinear uncertain systems. However, it provides some drawbacks linked to chattering, due to the higher needed switching gain in the case of large uncertainties. In order to reduce this gain, artificial neural network (ANN) is used for the estimation of model unknown parts and hence enable a lower switching gain to be used. The ANN is turned online; therefore, we cannot implement this control law in real time because of the time-varying delay generated by the ANN in online learning. This work is a proposal to compensate this delay so as to ensure the control in real time and maintain the advantage of online learning; we propose to convert the original system to a system with time-varying input delay which will practically represent the online learning time. We use Lyapunov-Krasovskii functional to prove the stability of the closed loop systems, and the effectiveness of the designed method is illustrated in simulations by the comparison with integral sliding mode control (ISMC).

A fuzzy neural network sliding mode controller for vibration suppression in robotically assisted minimally invasive surgery.

It is very important for robotically assisted minimally invasive surgery to achieve a high-precision and smooth motion control. However, the surgical instrument tip will exhibit vibration caused by nonlinear friction and unmodeled dynamics, especially when the surgical robot system is attempting low-speed, fine motion. A fuzzy neural network sliding mode controller (FNNSMC) is proposed to suppress vibration of the surgical robotic system. Nonlinear friction and modeling uncertainties are compensated by a Stribeck model, a radial basis function (RBF) neural network and a fuzzy system, respectively. Simulations and experiments were performed on a 3 degree-of-freedom (DOF) minimally invasive surgical robot. The results demonstrate that the FNNSMC is effective and can suppress vibrations at the surgical instrument tip. The proposed FNNSMC can provide a robust performance and suppress the vibrations at the surgical instrument tip, which can enhance the quality and security of surgical procedures.

Adaptive neural network sliding mode control for steer-by-wire-based vehicle stability control

This study develops a novel vehicle stability control (VSC) scheme using adaptive neural network sliding mode control technique for Steer-by-Wire (SbW) equipped vehicles. The VSC scheme is designed in two stages, i.e., the upper and lower level control stages. An adaptive sliding mode yaw rate controller is first proposed as the upper one to design the compensated steering angle for enabling the actual yaw rate to closely follow the desired one. Then, in the implementation of the yaw control system, the developed steering controller consists of a nominal control and a terminal sliding mode compensator where a radial basis function neural network (RBFNN) is adopted to adaptively learn the uncertainty bound in the Lyapunov sense such that the actual front wheel steering angle can be driven to track the commanded angle in a finite time. The proposed novel stability control scheme possesses the following prominent superiorities over the existing ones: (i) No prior parameter information on the vehicle and tyre dynamics is required in stability control, which greatly reduces the complexity of the stability control structure. (ii) The robust stability control performance against parameter variations and road disturbances is obtained by means of ensuring the good tracking performance of yaw rate and steering angle and the strong robustness with respect to large and nonlinear system uncertainties. Simulation results are demonstrated to verify the superior control performance of the proposed VSC scheme.

RBF neural network-based sliding mode vibration control of a flexible cantilever plate using laser displacement measurement

This paper is concerned with active vibration control of a flexible piezoelectric cantilever plate using a nonlinear radial basis neural network sliding mode control (RBFNN-SMC) algorithm and laser displacement measurement. In order to decouple the low-frequency vibration signals of the bending and torsional modes on measurement, two laser displacement sensors are used. The decoupling method is provided. A hyperbolic tangent function is used instead of the sign function, and the chattering phenomenon is alleviated. Also, the RBFNN is utilized to adjust the switching control gain adaptively to balance the chattering phenomenon and the control effect. The controllers for bending and torsional modes are designed independently. Experimental setup of the flexible piezoelectric cantilever plate with two laser displacement sensors is constructed. Experiments on vibration measurement and control are conducted by using the decoupling method and the designed controller, compared with the classical proportional and derivative (PD) control algorithm. The experimental results demonstrate that the proposed method can decouple the low-frequency bending and torsional vibration signals on measurement. Furthermore, the designed nonlinear RBFNN-SMC can suppress both the bending and torsional vibrations more quickly than the traditional linear PD controller, especially for the small amplitude residual vibration.

Adaptive wavelet network sliding mode control for a photovoltaic-pumping system

This paper presents an adaptive wavelet network sliding mode control (AWNSMC) scheme to track the maximum power point tracking in photovoltaic (PV) pumping systems. For the best use, the PV-generator must operate at its maximum power point. The proposed approach combines the advantages of the wavelet network (WN) and those of sliding mode control (SMC) to adjust the duty cycle control law in order to satisfy reaching conditions without producing chattering phenomenon. Indeed, due to its ability to approximate non-linear function and its fast convergence, a WN system is used to approximate the system uncertainties that are incorporated in the equivalent control component; as a result, it enables a lower switching gain to be used in the robust SMC term. The effectiveness of the designed AWNSMC is illustrated in simulations by a comparison with both standard SMC and neural network SMC techniques.

Design and implementation of a robust FNN-based adaptive sliding-mode controller for pneumatic actuator systems

An adaptive Fourier neural network sliding mode controller with H∞ tracking performance (AFNN-SMC+ H∞) is applied for a Pneumatic actuator system (PAS) to overcome time-varying nonlinear dynamics and external disturbances. Benefiting from the use of orthogonal Fourier basis function, the proposed AFNN has fast estimated convergence speed; also, because AFNN has unique solution, it can avoid falling into the local minimum. The architecture of AFNN can also easily be determined by its clear physical meaning of the neurons. To attenuate the vibration of proportional directional control valve and the adaptive approximation error, the H∞ tracking design technique is incorporated into the proposed AFNN-SMC. Finally, practical experiments are successfully implemented in position regulation, trajectory tracking, and velocity control of the PAS, which illustrates the effectiveness of the proposed controller.

Finite time convergent wavelet neural network sliding mode control guidance law with impact angle constraint

This paper presents a novel guidance law to intercept non-maneuvering targets with impact angle and lateral acceleration command constraints. Firstly, we formulate the impact angle control to track the desired line-of-sight (LOS) angle, which is achieved by selecting the missile's lateral acceleration to enforce the sliding mode on a sliding surface at impact time. Secondly, we use the Lyapunov stability theory to prove the stability and finite time convergence of the proposed nonlinear sliding surface. Thirdly, we introduce the wavelet neural network (WNN) to adaptively update the additional control command and reduce the high-frequency chattering of sliding mode control (SMC). The proposed guidance law, denoted WNNSMC guidance law with impact angle constraint, combines the SMC methodology with WNN to improve the robustness and reduce the chattering of the system. Finally, numerical simulations are performed to demonstrate the validity and effectiveness of the WNNSMC guidance law.

Intelligent Chaos Synchronization of Fractional Order Systems via Mean-Based Slide Mode Controller

In this paper, a new mean-based adaptive fuzzy neural network sliding mode control is developed to perform the chaos synchronization among the master–slave fractional order uncertain systems. The mean-based expansion is adopted to replace the traditional Taylor expansion to transform a nonlinear function into a partially linear form for the linearization of nonlinear systems. In comparison with the traditional Taylor method, the proposed mean-based method can estimate the first-order derivative term on the identifier model, which will somehow alleviate the computational burden. Based on the learning algorithms, the adaptive laws and control laws can be tuned on-line to synchronize the master–slave fractional order uncertain systems. Furthermore, the stability of the closed-loop system can not only be assured but the synchronization deviation of external perturbation can also be alleviated. Finally, simulation examples are illustrated to demonstrate the feasibility and the synchronization performance of this new approach.

Dynamic output feedback RBF neural network sliding mode control for robust tracking and model following

A dynamic output feedback radial basis function (RBF) neural network sliding mode control (SMC) scheme is proposed to realize the problem of robust tracking and model following for a class of uncertain time-delay systems. The algorithm is based on dynamic output feedback SMC, RBF neural network and adaptive control. The design of sliding surface and the existence of sliding mode have been addressed. The proposed robust tracking controller guarantees the stability of overall closed-loop system and achieves zero-tracking error in the presence of state delays, input delays, time-varying parameter uncertainties and external disturbances. Moreover, the knowledge of upper bound of uncertainties is not required, and chattering phenomenon is eliminated. Both theoretical analysis and illustrative examples demonstrate the validity of the proposed scheme.