MANFIS Approach for Path Planning and Obstacle Avoidance for Mobile Robot Navigation

Abstract

Path planning and obstacle avoidance are very crucial issues for an Autonomous mobile robot. In this research paper an intelligent hybrid approach MANFIS (Multiple Adaptive Neuro-Fuzzy Inference system) has been implemented for mobile robot navigation. The adaptive neuro-fuzzy inference system (ANFIS) has taken the advantages of expert knowledge of fuzzy inference system and learning capability of artificial neural network. The inputs to the MANFIS controller include the front obstacle distance, the left obstacle distance, the right obstacle distance and the target angle and outputs from the controller are left wheel velocity and right wheel velocity of the mobile robot. In order to validate the proposed hybrid technique a series of simulation experiments using MATLAB were performed and it was found that the proposed navigational controller is capable to avoid obstacle and reach the destination successfully. The experimental results also have been compared with simulation results to prove the authenticity of the developed navigational controller MANFIS.