Dynamic path planning algorithm in mobile robot navigation

Abstract

Mobile Robot Navigation is an advanced technique where static, dynamic, known and unknown environment is involved. In this research, Genetic Algorithm (GA) is used to assist mobile robot to move, identify the obstacles in the environment, learn the environment and reach the desired goal in an unknown and unrecognized environment. This study is focused on exploring the algorithm that avoids acute obstacles in the environment. In the event of mobile robot encountering any dynamic obstacles when travelling from the starting position to the desired goal according to the optimum collision free path determined by the controller, the controller is capable of re-planning the new optimum collision free path. MATLAB simulation is developed to verify and validate the algorithm before they are real time implemented on Team AmigoBotℒ robot. The results obtained from both simulation and actual application confirmed the flexibility and robustness of the controllers designed in path planning.