Self-tuning PID controller and its application to a DD robot manipulator

In this paper, a self-tuning PID Controller is proposed. This PID controller consists of two paralleled loops: one is an adaptive PID controller, and the other one is a neural PID controller. The adaptive PID controller is designed by adaptive pole placement method, aimed to control the linear part of a plant in main. The gains of the adaptive PID controller are obtained on the basis of the relationship between the PID gains and the closed-loop poles. As for the neural PID controller, a neural network is adopted to adjust the PID gains. With the proposed self-tuning PID control scheme, a two-link DD (Direct Drive) robot manipulator is controlled by two adaptive pole-placement PID controllers. Then a paralleled neural PID controller, which supplies other two pairs of PID gains for the two arms respectively, provides the necessary compensation for the nonlinear dynamics and the interference of the DD robot manipulator. Experimental results show that the proposed self-tuning control scheme can achieve very good performance.