Obstacle avoidance strategy for mobile robot based on improved coordinating potential field with genetic algorithm

To overcome the problems during navigation of mobile robots in dynamic environment using the traditional artificial potential field (APF) method, a novel improved method called coordinating potential field (CPF) was proposed. The local potential field was constructed by using local subgoals, which was obtained by updating dynamic rolling window. The questions of local minima, oscillation between multiple obstacles and real-time dynamic obstacles avoidance were solved. At last multi-objective parameter optimization was implemented by using adaptive genetic algorithm. Simulation results indicate that the strategy is feasible and practicable.