Mobile Robot Path Planning Using Genetic Algorithm Global Path PLANNING and Potential Field Path Adjusting

Abstract

The purpose of this thesis was to develop an algorithm which solves the path planning problem for a two-wheeled mobile robot. The algorithm included multiple robots within it. The algorithm was developed by first identifying the problem at hand, expanding on it and then designing features which solved each part of the problem. The motivation behind this study is that the path planning algorithm could be applied to real world applications once perfected to a certain degree. The algorithm was composed of two main sub-algorithms: an off-line component and an on-line one. The off-line component was a global path planner which used a genetic algorithm to find the optimal path between the source point and the goal point while only accounting for static obstacles. The second component was a local path planner utilizing artificial potential fields in order to adjust the path at hand to account for any dynamic objects. Multiple configurations were tested and evaluated. The algorithm was compared to an Ant Colony Optimization algorithm and it proved to be more efficient with respect to the maps used in the study. The algorithm was evaluated for advantages and limitations and future study aspects were identified. Future works included: expanding the algorithm to allow for more robots, implementing the algorithm to a real life application and recreating miscellaneous situations to test the flexibility of the algorithm.