Position control of a flexible manipulator using a new nonlinear self-tuning PID controller

Abstract

In this paper, a new nonlinear self-tuning PID controller &#x0028 NSPIDC &#x0029 is proposed to control the joint position and link deflection of a flexible-link manipulator &#x0028 FLM &#x0029 while it is subjected to carry different payloads. Since, payload is a critical parameter of the FLM whose variation greatly influences the controller performance. The proposed controller guarantees stability under change in payload by attenuating the non-modeled higher order dynamics using a new nonlinear autoregressive moving average with exogenous-input &#x0028 NARMAX &#x0029 model of the FLM. The parameters of the FLM are identified on-line using recursive least square &#x0028 RLS &#x0029 algorithm and using minimum variance control &#x0028 MVC &#x0029 laws the control parameters are updated in real-time. This proposed NSPID controller has been implemented in real-time on an experimental set-up. The joint tracking and link deflection performances of the proposed adaptive controller are compared with that of a popular direct adaptive controller &#x0028 DAC &#x0029 . From the obtained results, it is confirmed that the proposed controller exhibits improved performance over the DAC both in terms of accurate position tracking and quick damping of link deflections when subjected to variable payloads.