Neural Network Control Scheme Research Articles

Robust control based on adaptive neural network for Rotary inverted pendulum with oscillation compensation

A new stable adaptive neural network (ANN) control scheme for the Furuta pendulum, as a two-degree-of-freedom underactuated nonlinear system, is proposed in this paper. This approach aims to address the control problem of the Furuta pendulum in the steady state and also in the presence of external disturbances. The adaptive classical control laws such as e-modification present some limitations in particular when oscillations are presented in the input. To avoid this problem, two ANNs are implemented using filtered tracking error in the control loop. The first one is a single hidden layer network, used to approximate the equivalent control online, and the second is the feed-forward network, used to minimize the oscillations. The goal of the control is to bring the pendulum close to the upright position in the presence of the various uncertainties and being able to compensate oscillations and external disturbances. The main purpose of the second ANN is to minimize the chattering phenomenon and response time by finding the optimal control input signal, which also leads to the reduction of energy consumption. The learning algorithms of the two ANNs are obtained using the direct Lyapunov stability method. The simulation results are given to highlight the performances of the proposed control scheme.

Adaptive recursive sliding-mode dynamic surface and its event-triggered control of uncertain non-affine systems

An adaptive neural network control strategy and its event-triggered controller are designed for non-affine pure-feedback uncertain systems based on nonlinear gains and recursive sliding-mode dynamic surfaces. A kind of nonlinear gain functions is introduced into the traditional framework of dynamic surface control (DSC) with recursive sliding-mode to make a compromise between control accuracy and transient performance. Radial basis function (RBF) neural networks (NNs) are adopted to approximate unknown functions at each step, and novel adaptive update laws with leakage terms of σ-modification is constructed. To reduce the action number of the actuator, an extended control strategy in an event-triggered manner is proposed. By the Lyapunov function, it is proven that both of the two control strategies can force the tracking error arbitrarily small and guarantee all the signals in the closed-loop system uniformly ultimately bounded. Finally, simulation results are provided to verify the effectiveness of the proposed control strategy.

Adaptive Tracking Control of Cooperative Robot Manipulators With Markovian Switched Couplings

Many cooperative robotic systems have not only modeling heterogeneity and uncertainty but also switched couplings, causing control difficulties. Here, we develop a neural network adaptive control framework for cooperative robot manipulators with unknown Euler–Lagrange dynamics and Markovian switched couplings. Second-order Markovian switching networks are used for modeling such cooperative robotic systems, which admit a hybrid neural network control with a desired tracking performance. The hybrid neural network control scheme contains a distributed adaptive controller and a hybrid adaptation law, enabling learning in the closed-loop system. The position and velocity tracking errors are shown to be practically uniformly exponentially stable in the mean-square sense, respectively, guaranteeing the second-order practical tracking. The results also suggest that the neural weight evolves with practical convergence to the ideal, showing the effect of network structure on the adaptation capacity.

Precise Motion Control of Wafer Stages via Adaptive Neural Network and Fractional-Order Super-Twisting Algorithm

To obtain precise motion control of wafer stages, an adaptive neural network and fractional-order super-twisting control strategy is proposed. Based on sliding mode control (SMC), the proposed controller aims to address two challenges in SMC: 1) reducing the chattering phenomenon, and 2) attenuating the influence of model uncertainties and disturbances. For the first challenge, a fractional-order terminal sliding mode surface and a super-twisting algorithm are integrated into the SMC design. To attenuate uncertainties and disturbances, an add-on control structure based on the radial basis function (RBF) neural network is introduced. Stability analysis of the closed-loop control system is provided. Finally, experiments on a wafer stage testbed system are conducted, which proves that the proposed controller can robustly improve the tracking performance in the presence of uncertainties and disturbances compared to conventional and previous controllers.

Neural Network-Based Closed-Loop Deep Brain Stimulation for Modulation of Pathological Oscillation in Parkinson’s Disease

Aiming at the problem that the Proportional-Integral-Derivative (PID) control strategy needs to readjust controller parameters for different Parkinson's disease (PD) states. This work proposes an improved control strategy that considers an artificial neural network control scheme. A backpropagation neural network (BPNN) controller is designed to solve the above problem and further to improve the performance of the closed-loop control strategy. The training data set of the BPNN controller is obtained by controlling eight different PD states (PD a - PD h ) by the PID controller and the BPNN controller is trained by the training data set to obtain a set of optimal weights. By modulating other different PD states (e.g. PD1 - PD3), the effectiveness of the PID-structure controller and BPNN controller are compared. We find that the BPNN controller can modulate different PD states without changing the controller parameters and reduce energy expenditure by 58.26%. This work is helpful for the design of more effective closed-loop deep brain stimulation (DBS) systems for clinical applications and provides a framework for the further development of closed-loop DBS.

Adaptive Finite-Time Neural Control for a Class of Stochastic Nonlinear Systems With Known Hysteresis

This manuscript considers the finite time adaptive neural tracking control problem for a class of the nonstrict-feedback stochastic nonlinear systems with Bouc-Wen hysteresis input. During the design process, a Bouc-Wen model is first adopted to obtain the input hysteresis phenomenon. By fusion with the backstepping technique and the neural network approximation capability, the unknown nonlinearities are coped with, a constructive finite time adaptive neural network control strategy is proposed. Furthermore, the finite-time mean square stability of stochastic nonlinear systems is proved, and at last the effectiveness of the proposed control strategy is validated by the simulations.

Low Delay Control Algorithm of Robot Arm for Minimally Invasive Medical Surgery

Minimally invasive surgical robots have received more and more attention from medical patients because of their higher surgical accuracy and higher safety than doctors. Minimally invasive surgery is rapidly revolutionizing the treatment of traditional surgery. In order to solve the problem that the surgical robot has a redundant degree of freedom, which makes the kinematics solution more complicated, this paper analyzes the kinematics of the coordinate system block. Aiming at the problem that the strategy search algorithm needs to re-learn when the target pose changes, a convolutional neural network control strategy is studied and constructed. By designing the structure of the convolutional neural network visual layer and motor control layer, the loss function and sampling of the training process are established. Aiming at the problem of long training time of convolutional neural network, an effective pre-training method is proposed to shorten the training time of the neural network. At the same time, the effectiveness of the above method and the end-to-end control of the convolutional neural network strategy are verified through simulation experiments. The physical structure of the manipulator body is analyzed, and the forward and inverse kinematic equations of the manipulator are established by the D-H method. Monte Carlo method was used to analyze the working space of the manipulator, and low-latency control and simulation experiments were carried out on the movement trajectory of the manipulator in joint space and Cartesian space. The results show that the low-latency control algorithm in this paper is effective to control the mechanical arm of the minimally invasive medical surgery robot.

Adaptive Neural Network Control for Active Suspension Systems With Time-Varying Vertical Displacement and Speed Constraints

In this paper, an adaptive neural network (NN) control scheme is proposed for a quarter-car model, which is the active suspension system (ASS) with the time-varying vertical displacement and speed constraints and unknown mass of car body. The NNs are used to approximate the unknown mass of car body. It is commonly known that the stability and security of the ASSs will be weakened when the constraints are violated. Thus, the control problem of the time-varying vertical displacement and speed constraints for the quarter-car ASSs is a very important task because of the demand of the handing safety. The time-varying barrier Lyapunov functions are used to guarantee the constraints of the vertical displacement not violated, and it can prove the stability of the closed-loop system. Finally, a simulation example for the ASSs is employed to show the feasibility and rationality of the proposed approach.

Performance enhancement of Uninterruptible Power Supply inverter through Neural Network control strategy

Performance enhancement of Uninterruptible Power Supply inverter through Neural Network control strategy

Maximum power point tracking in wind energy conversion system using radial basis function based neural network control strategy

In this paper, a neural network tuned controller for maximum power point tracking (MPPT) in wind energy conversion system (WECS) is proposed. This technique utilizes radial basis function based neural networks (RBF-NN) and tracks the MPP using duty cycle control. The WECS is based on permanent magnet synchronous generator (PMSG). The rectifier output voltage and power are measured and fed into an RBF-NN controller. The MPPT algorithm controls the duty cycle of a DC-DC boost converter to extract the maximum power. The output of the boost converter is tied to the grid using a voltage source inverter (VSI). Unlike conventional methods, MPPT using proposed method does not require knowledge of wind turbine power characteristics, thus minimizes the need for various measuring instruments. The method implemented is also compared with other commonly used MPPT techniques like fuzzy logic control (FLC), perturb & observe (P&O), and back-propagation (BP) based NN. The proposed system provides better results as compared with other relevant results available in the literature. The proposed method is implemented using MATLAB/Simulink and then validated in real-time using digital simulation hardware, OPAL-RT 4510.

Adaptive Neural Backstepping Control Design for A Class of Nonsmooth Nonlinear Systems

This paper investigates the problems of backstepping control for a class of nonsmooth nonlinear systems. With the help of set-valued functions (or maps) and set-valued derivatives, a new Lyapunov criterion is developed for nonsmooth systems. The concept of semi-globally uniformly ultimately bounded (SGUUB) stability that is widely used for smooth nonlinear systems in lower triangular form is, for the first time, applied to the nonsmooth case, and such a concept provides a theory foundation for the subsequent backstepping control design. Then, two types of continuous controllers are designed based on the backstepping technique and adaptive neural network (NN) algorithm. In the light of Cellina approximate selection theorem and smooth approximation theorem for Lipschitz functions, the system under investigation is first transformed into an equivalent model. Then, the first type of controller can be efficiently designed by using the traditional backstepping technique. On the other hand, when the unknown uncertainty is taken into account in the equivalent systems, another type of controller is designed based on adaptive NN control strategy. It is proved that in both cases, all the closed-loop signals are SGUUB by using our proposed controllers. Finally, a numerical example with two types of controllers is supplied to show the availability of the provided control schemes.

Disturbance Observer-Based Neural Network Control of Cooperative Multiple Manipulators With Input Saturation.

In this paper, the complex problems of internal forces and position control are studied simultaneously and a disturbance observer-based radial basis function neural network (RBFNN) control scheme is proposed to: 1) estimate the unknown parameters accurately; 2) approximate the disturbance experienced by the system due to input saturation; and 3) simultaneously improve the robustness of the system. More specifically, the proposed scheme utilizes disturbance observers, neural network (NN) collaborative control with an adaptive law, and full state feedback. Utilizing Lyapunov stability principles, it is shown that semiglobally uniformly bounded stability is guaranteed for all controlled signals of the closed-loop system. The effectiveness of the proposed controller as predicted by the theoretical analysis is verified by comparative experimental studies.

Neuro-Sliding Control for Underwater ROV's Subject to Unknown Disturbances.

Proposed in this paper is a model-free and chattering-free second order sliding mode control (2nd-SMC) in combination with a backpropagation neural network (BP-NN) control scheme for underwater vehicles to deal with external disturbances (i.e., ocean currents) and parameter variations caused, for instance, by the continuous interchange of tools. The compound controller, here called the neuro-sliding control (NSC), takes advantage of the 2nd-SMC robustness and fast response to drive the position tracking error to zero. Simultaneously, the BP-NN contributes with its capability to estimate and to compensate online the hydrodynamic variations of the vehicle. When a change in the vehicle’s hydrodynamics occurs, the 2nd-SMC may no longer be able to compensate for the variations since its feedback gains are tuned for a different condition; thus, in order to preserve the desired performance, it is necessary to re-tune the feedback gains, which a cumbersome and time consuming task. To solve this, a viable choice is to implement a BP-NN control scheme along with the 2nd-SMC that adds or removes energy from the system according to the current condition it is in, in order to keep, or even improve, its performance. The effectiveness of the proposed compound controller was supported by experiments carried out on a mini-ROV.

Fuzzy Neural Network Control of Thermostatically Controlled Loads for Demand-Side Frequency Regulation

In this paper, a fuzzy neural network controller for regulating demand-side thermostatically controlled loads (TCLs) is designed with the aim of stabilizing the frequency of the smart grid. Specifically, the balance between power supply and demand is achieved by tracking the automatic generation control (AGC) signal in an electric power system. The particle swarm optimization (PSO) and error back propagation (BP) algorithms are used to optimize the control parameters and consequently reduce the tracking errors. The fuzzy neural network can be applied to solve load control problems in power systems, since its self-learning and associative storage functions can deal with the highly nonlinear relationship between input and output. Simulation results show the advantage of the fuzzy neural network control scheme in terms of frequency regulation error and consumer comfort.

Data-driven-based adaptive fuzzy neural network control for the antimony flotation plant

Due to the unknown system structure of the froth flotation process and frequent fluctuations in production conditions, design of control strategy is a challenging problem. As a result, manual operation is still widely applied in practice by observing froth image features. However, since the manual observation is subjective and the production conditions are time-varying, the manual operation cannot make decisions quickly and accurately. In this paper, a data-driven-based adaptive fuzzy neural network control strategy is developed to implement the automatic control of the antimony flotation process. The strategy is composed of fuzzy neural network (FNN) controllers, a data-driven model, and an on-line adaptive algorithm. The FNN is constructed to derive the control laws of the reagent dosages. The parameters of the FNN controllers are tuned by gradient descent algorithm. To obtain the real-time error feedback information, the data-driven model is established, which integrates the long short term memory (LSTM) network and radial basis function neural network (RBFNN). The LSTM network is utilized as a primary model, and the RBFNN is used as an error compensation model. To handle the challenges of the frequent fluctuations in the production conditions, the on-line adaptive algorithm is proposed to tune the parameters of the FNN controllers. Simulations and experiments are carried out in a real-world antimony flotation plant in China. The results demonstrate that the proposed adaptive fuzzy neural network control strategy produces better control performance than the other two existing methods.

Adaptive finite‐time neural network control for redundant parallel manipulators

AbstractThis paper reports our study of an adaptive finite‐time neural network control scheme for redundant parallel manipulators. The proposed controller is based on fully tuned radial basis function neural networks (RBFNs), nonsingular fast terminal sliding mode control (NFTSMC), and nonlinearities in output feedback. RBFNs with a fully online update on the output weights and both the centers and variances of Gaussian functions are used to estimate system uncertainties and disturbances. The proposed approach possesses several advantages over other existing approaches like robustness, rapid response, nonsingularity, higher precision, finite‐time convergence, and better tracking control performance. The stability of the parallel manipulators is ensured by the Lyapunov theory. Finally, simulation results validate the effectiveness of the proposed methodology.

Minimal learning parameters-based adaptive neural control for vehicle active suspensions with input saturation

In this paper, a neural network (NN) control method is developed for nonlinear quarter vehicle active suspension systems (VASSs) which have the features of parameter uncertainties, input saturation and road disturbance, whose aim is to ensure safe driving and improve the ride comfort. The NNs are employed to approximate unkown nonlinear functions that the forming reason is uncertainties by caused varied sprung mass. When the output of actuator goes beyond its maximums, an NN control scheme combined with anti-saturation is proposed to handle this problem. Furthermore, an NN controller with the minimal learning parameters is constructed to ensure that the number of adaptive learning parameters and the burden computation are largely reduced for VASSs. Meanwhile, the control objectives of VASSs are proved based on the stability analysis. Finally, the effectiveness of designed scheme is demonstrated by an example.

A small‐gain approach for adaptive output‐feedback NN control of switched pure‐feedback nonlinear systems

SummaryThis paper investigates the problem of adaptive output‐feedback neural network (NN) control for a class of switched pure‐feedback uncertain nonlinear systems. A switched observer is first constructed to estimate the unmeasurable states. Next, with the help of an NN to approximate the unknown nonlinear terms, a switched small‐gain technique‐based adaptive output‐feedback NN control scheme is developed by exploiting the backstepping recursive design scheme, input‐to‐state stability analysis, the common Lyapunov function method, and the average dwell time (ADT) method. In the recursive design, the difficulty of constructing an overall Lyapunov function for the switched closed‐loop system is dealt with by decomposing the switched closed‐loop system into two interconnected switched systems and constructing two Lyapunov functions for two interconnected switched systems, respectively. The proposed controllers for individual subsystems guarantee that all signals in the closed‐loop system are semiglobally, uniformly, and ultimately bounded under a class of switching signals with ADT, and finally, two examples illustrate the effectiveness of theoretical results, which include a switched RLC circuit system.

Adaptive neural network auto-berthing control of marine ships

To solve the auto-berthing control problem for underactuated marine ships subject to unknown dynamics and external disturbances under the circumstance of the restricted waters in port, firstly, an additional control method is adopted to solve the underacruated problem. Secondly, a robust neural network (NN) adaptive approach based on the navigation dynamic deep-rooted information (DRI) is proposed to reconstruct the lumped uncertainties caused by unknown ship dynamics and external disturbances. Meanwhile, the dynamic surface control (DSC) and the minimum learning parameter (MLP) techniques are used to reduce the computational load of the adaptive NN control scheme. Considering the input saturation effects of control actuator (rudder, propeller, etc) and the coupling characteristics of uncertainties, this approach integrates neural network weights, approximation errors and external disturbance term as the composite uncertain parameters, which is estimated online by parameter adaptive technique. Finally, simulations are carried out on an underactuated model ship to verify the effectiveness of the proposed auto-berthing control scheme.

Fuzzy Tracking Control for a Class of Uncertain MIMO Nonlinear Systems With State Constraints

In this paper, an adaptive fuzzy neural network (FNN) control scheme is developed for a class of multiple-input and multiple-output (MIMO) nonlinear systems subject to unknown dynamics and state constraints. FNNs are used to approximate the unknown dynamics that comprises the effects of uncertain parameters and functions. Also, integral Lyapunov functions are introduced to address state constraints. A neural-network-based observer is designed to estimate the unmeasurable states. With state-feedback and output feedback tracking control, the stability of closed-loop system is guaranteed via Lyapunov’s stability theory. Two cases of simulations for MIMO systems with state constraints are conducted to verify the effectiveness of the proposed control.