PSO based adaptive learning Fuzzy Logic Controller for the Irobot Create robot

In real robot applications, the task of designing a fuzzy logic controller is complex enough essentially because the presence of many forms of noise and uncertainties. The robot while navigating has to control many variables to get the best result at the end of the task: best smooth trajectory, the guarantee of arrive to goal, lowest time, etc. We present in this paper a novel Particle Swarm Optimization based adaptive learning Fuzzy Logic Controller design for a motion planning task. The membership functions of a fuzzy controller are tuned instantanously using particle swarm optimization technique. The proposed architecture presented good results which were demonstrated on the robot “iRobot Create”.