Adaptive control of nonlinear uncertain systems using neural network

This article presents a new command-filtered composite adaptive neural control scheme for uncertain nonlinear systems. Compared with existing works, this approach focuses on achieving finite-time convergent composite adaptive control for the higher-order nonlinear system with unknown nonlinearities, parameter uncertainties, and external disturbances. First, radial basis function neural networks (NNs) are utilized to approximate the unknown functions of the considered uncertain nonlinear system. By constructing the prediction errors from the serial-parallel nonsmooth estimation models, the prediction errors and the tracking errors are fused to update the weights of the NNs. Afterward, the composite adaptive neural backstepping control scheme is proposed via nonsmooth command filter and adaptive disturbance estimation techniques. The proposed control scheme ensures that high-precision tracking performances and NN approximation performances can be achieved simultaneously. Meanwhile, it can avoid the singularity problem in the finite-time backstepping framework. Moreover, it is proved that all signals in the closed-loop control system can be convergent in finite time. Finally, simulation results are given to illustrate the effectiveness of the proposed control scheme.

In this paper; a new robust adaptive neural network controller (RANNC) is presented for a class of non-affine nonlinear systems in the presence of unknown nonlinearities and disturbances. Firstly, the existence of an ideal implicit feedback linearisation control (IFLC) is established based on implicit function theory. Using Taylor series expansion, it is shown that the control of non-affine nonlinear systems is equivalent to the control ofaffine nonlinear systems in the neighbourhood of the operating trajectory under mild conditions. Then, a robust adaptive neural network control scheme is presented for the transformed nonlinear systems by using neural networks as universal approximators for the unknown system nonlinearities. The proposed RANNC can guarantee that all the signals in the closed-loop system are bounded and the tracking error asymptotically converges to zero. Simulation studies on the control of a continuously stirred tank reactor (CSTR) system are used to show the effectiveness of the scheme.

Decentralized neural identification and control for uncertain nonlinear systems: Application to planar robot

This paper deals with an adaptive robust tracking control using a multilayer neural network (NN) for a class of nonlinear dynamic systems with unknown time varying state delays. Typical adaptive NN backstepping controllers for uncertain nonlinear systems with time-delay give rise to computation complexity caused by the the repeated derivatives of virtual controllers and nonlinear functions. Moreover, the combined techniques usually result in only uniformly ultimately bounded (UUB) stability caused by the inherent NN approximation error. This paper presents a control scheme that uses an integral sliding mode control as a feedback term and an adaptive neural controller as a feedforward term based on the desired compensation adaptive law (DCAL) technique. First, we develop a new DCAL formulation which avoids the explosion of complexity caused by the general NN backstepping scheme to compensate for nonlinear system uncertainties, bounded system disturbances, and unknown state time delays. Then, using a Lyapunov-Krasovskii (LK) functional, it is shown that the proposed controller renders the class of uncertain nonlinear time-delay systems asymptotically stable.

Adaptive disturbance rejection neural output feedback control of hydraulic manipulator systems

This paper describes a hybrid approach to the problem of controlling a class of nonlinear systems in the face of both unknown nonlinearities and unmodeled dynamics. In the proposed methodology, neural networks and direct adaptive control compensate for unknown nonlinear mappings while dynamic nonlinear damping provides robustness to unmodeled dynamics. To illustrate the theoretical development, the authors design a longitudinal autopilot for a simplified nonlinear missile aerodynamic model with input unmodeled dynamics. Simulation results demonstrate robustness in the face of this type of uncertainty.

Deterministic learning-based neural control for output-constrained strict-feedback nonlinear systems

Data-based adaptive neural network optimal output feedback control for nonlinear systems with actuator saturation

In this article, the adaptive event-triggered fault-tolerant control (FTC) issue of uncertain nonlinear systems suffering from sensor and actuator faults as well as external disturbances is studied. A disturbance observer (DO) is constructed to compensate for external disturbances and approximation errors generated by neural networks (NNs). Then, switching event-triggered control and command filtering techniques are introduced to balance the communication frequency and tracking performance of the controlled systems while avoiding “explosion of complexity” issue caused by iterative derivations of virtual controllers. Furthermore, compensation signals are designed to eliminate filtered errors. Finally, it is proven that the developed FTC scheme can esure that all signals are bounded and free from Zeno phenomenon. Simulation results of a spring damping mechanical system verifies the merits of the designed control algorithm.

An adaptive neural network control scheme is developed for a class of nonlinear systems in the strict-feedback form. Compared with the existing approaches, the main advantage is that the developed scheme can be implemented by utilizing only one neural network approximator. Thus, the designed controller structure is simplified. In addition, less neural network can reduce the running cost in practical application. The developed neural network control scheme can achieve that all the signals of the closed-loop system are uniformly bounded and the tracking errors converge to an arbitrary small neighborhood around zero by selecting suitably design parameters.

In this paper, we present a neural adaptive control scheme for active vibration suppression of a composite aircraft fin tip. The mathematical model of a composite aircraft fin tip is derived using the finite element approach. The finite element model is updated experimentally to reflect the natural frequencies and mode shapes very accurately. Piezo-electric actuators and sensors are placed at optimal locations such that the vibration suppression is a maximum. Model-reference direct adaptive neural network control scheme is proposed to force the vibration level within the minimum acceptable limit. In this scheme, Gaussian neural network with linear filters is used to approximate the inverse dynamics of the system and the parameters of the neural controller are estimated using Lyapunov based update law. In order to reduce the computational burden, which is critical for real-time applications, the number of hidden neurons is also estimated in the proposed scheme. The global asymptotic stability of the overall system is ensured using the principles of Lyapunov approach. Simulation studies are carried-out using sinusoidal force functions of varying frequency. Experimental results show that the proposed neural adaptive control scheme is capable of providing significant vibration suppression in the multiple bending modes of interest. The performance of the proposed scheme is better than the H(infinity) control scheme.

This article focuses on the asymptotic tracking control problem for uncertain nonlinear systems subject to both multiple disturbances and parametric uncertainties. To address this issue, a parameter adaptation law is synthesized to deal with the parametric uncertainties, and an adaptive-gain disturbance estimator (ADE) is constructed to estimate the mismatched and matched disturbances, and compensate them in feedforward channels, which eliminates the impact of disturbances on tracking performance. Meanwhile, an updated law for estimator gain driven by the estimation errors is utilized in the ADE when facing unknown upper bounds of disturbances, which reduces the conservatism of estimator gain selection and is beneficial to practical implementation. Based on the parameter adaption technique and the presented ADE approach, a composite controller is proposed to ensure an excellent asymptotic output tracking performance. The stability analysis shows the proposed controller can attain asymptotic tracking performance in the presence of both time-variant disturbances and parametric uncertainties. Comparative simulation results of the application to a robot manipulator reveal the validity of the developed approach.

Adaptive neural control for uncertain stochastic nonlinear strict-feedback systems with time-varying delays: A Razumikhin functional method

Fuzzy rule-based neural appointed-time control for uncertain nonlinear systems with aperiodic samplings

Neural network-based adaptive fault-tolerant control for nonlinear systems with unknown backlash-like hysteresis and unmodeled dynamics