Self-Tuning PID Controller for Serial Robots with Real-Time Functional Optimization

This paper presents an interesting application for Robust Extremum Seeking Algorithm (RESA) to a self-tuning PID controller for serial robots. The theoretical framework is presented to guarantee the stability and convergence of this proposal. The main advantage of this method is that it has an easy-to-implement the structure and does not require strong mathematical conditions allowing minimize a measurable convex functional in real-time which is selected to reduce the error norm and its derivative however there is the possibility of choosing it in another way. To validate the methodology we applied it to develop self-tuning Proportional-Integral-Derivative (PID) controller for a Selective Compliant Assembly Robot Arm (SCARA) robot.