Fuzzy controller for flexible-link robot arm by reduced-order techniques

Abstract

The design and analysis of a large-scale control system should be based on the best available knowledge instead of the simplest available model when treating uncertainties in the system. Therefore, a large-scale system is better treated by knowledge-based methods such as fuzzy logic, neural networks, expert systems, etc. This paper concentrates on fuzzy logic using the singular perturbation approach for flexible-link robot arm control. To reduce the spillover effect, we will introduce a singular perturbation approach to derive the slow and fast subsystems. A composite control design is adopted. Therefore, a two-time scale fuzzy logic controller will be applied to the system. The fast-subsystem controller will damp out the vibration of the flexible structure by an optimal control method. Hence, the slow-subsystem fuzzy controller dominates the trajectory tracking. We guarantee the stability of the internal dynamics by adding a boundary-layer correction based on singular perturbations. Various case studies are given to verify the control algorithm. It appears that the fuzzy control method is quite useful in terms of reliability and robustness.