Robust control for nonlinear motor-mechanism coupling system using wavelet neural network

Nonlinear motor-mechanism coupling tank gun control system based on adaptive radial basis function neural network optimised computed torque control

This study investigates the spatial pointing control of a motor-mechanism coupling tank gun. The tank gun control system (TGCS) is driven and stabilised by the motor servo system. However, complicated nonlinearities in the TGCS are inevitable, such as friction, parameter uncertainty, and modelling errors. To solve this problem, the TGCS is regarded as a coupling system composed of mechanical, motor, and control systems. Accordingly, the mechanical and motor models of the marching tank gun are developed first in this paper. The motor-mechanism coupling dynamics model is established based on the principle of equivalent torque. On this basis, a computed torque controller, whose uncertainty was estimated using a radial basis function neural network (RBFNN), is constructed. A modified adaptive algorithm is used to estimate the weights of the RBFNN, and the estimation error of the uncertain observer is compensated by a compensation controller. Simulation results under different conditions validated the effectiveness of the proposed control system, revealing that the proposed control system has good tracking accuracy, strong adaptability, and robustness.

A probabilistic wavelet system for stochastic and incomplete data-based modeling

A probabilistic wavelet system (PWS) is proposed to model the unknown dynamic system with stochastic and incomplete data. When compared with the traditional wavelet system, the PWS uses a novel three-domain wavelet function to make a balance among the probability, time, and frequency domains, which achieves a robust modeling performance with poor data information. The definition, transformation, multiple-resolution analysis, and implementation of the PWS are presented to construct the whole theoretical framework. Simulation studies show that the performance of the proposed PWS is superior to the traditional one in a stochastic and incomplete data environment.