Fuzzy Behavior-Based Control for the Obstacle Avoidance of Multi-link Manipulators

Abstract

The behavior based approach has been actively used in many applications of intelligent robots due to the advantages of dividing the control system according to the task achieving behaviors over the conventional method in which the division is based on functions. One important application that had been done is for a mobile robot to reach a target while avoiding obstacles. The objective of this paper is for a multi-link manipulator to reach a target while avoiding obstacles by using a fuzzy behavior-based control approach. The control system that had been applied to the mobile robot in the previous work, is modified to suit to the manipulator. Fuzzy behavior elements are trained by a genetic algorithm. An additional component is also introduced in order to overcome the gravitational effect. Simulation results show that the manipulator reaches the target with an acceptable solution.