Control of a Flexible Robot Arm using a Simplified Fuzzy Controller

Abstract

A flexible robot arm is a distributed system per se. Its dynamics are very complicated and coupled with the non-minimum phase nature due to the non-collocated construction of the sensor and actuator. This gives rise to difficulty in the control of a flexible arm. In particular, the control of a flexible arm usually suffers from control spillover and observation spillover due to the use of a linear and approximate model. The robustness and reliability of the fuzzy control have been demonstrated in many applications, particularly, it is perfect for a nonlinear system without knowing the exact system model. However, a fuzzy control usually needs a lot of computation time. In order to alleviate this restraint, a simplified fuzzy controller is developed for real-time control of a flexible robot arm. Furthermore, the self-organizing control based on the simplified fuzzy controller is also developed. The simulation results show that the simplified fuzzy control can achieve the desired performance and the computation time is less than 10 m s so that the real-time control is possible.