Autonomous robot navigation system using a novel value encoded genetic algorithm

This paper describes the development of a genetic algorithm (GA) based path-planning software for local obstacle avoidance. The GA uses a novel encoding technique, which was developed to optimize the information content of the GA structure. Simulation results were used to further optimize the developed software and determine its optimum field of operation. The results show that the GA finds valid solutions to the path-planning problem within reasonable time and can therefore be used for real world applications.