Improved Ant Colony Algorithm based on cellular Automata for obstacle avoidance in robot soccer

Abstract

This paper introduces the evolvement rule of Cellular Automata (CA) into robot soccer competition and combines CA with Ant Colony Algorithm, then utilizes them to help soccer robot avoid obstacles. In order to avoid local convergence of the algorithm, we present a pheromone-diffusion mechanism and a target distance stimulating factor to increase the guidance quality of the ant colony in optimization, so that we improve the algorithm's accuracy and convergence speed. Then, we utilize arc spline to smooth the feasible path. Finally, simulation results prove the effectiveness and feasibility of the algorithm.