Control system design for ionic polymer metal composite using a single neuron based adaptive PID approach

Owing to simple structure and strong robustness characteristics, the traditional (proportional-integral-derivative) PID controllers have been widely used in designing control systems of industrial process. However, due to complex nonlinear properties in some dynamics plants, it is difficult to obtain accurate mathematical models, and vulnerable to the object and environment. So, it makes difficult to setting controller parameters, especially no online self-tuning. Addressing an ionic polymer metal composite (IPMC) artificial muscle with high nonlinear properties and model uncertainties, an IPMC position tracking control system based on single neuron adaptive-PID control approach is proposed by using neural network self-learning and nonlinear mapping ability. The designed system not only can achieve position tracking and online self-tuning, but also guarantee robust stability in the presence of effect of uncertainties the effectiveness of the proposed method is confirmed by simulation results.