A Collision Predict Based Local Path Planning of Mobile Robots

Abstract

Based on the velocity of obstacles and its distribution, the position of the dangerous region (potential collision region) has been predicted. So that the dangerous region would be expanded based on the distribution of its position to ensure the security. By finding the best “Free Road” and estimate the “danger”, a data reduction technique is employed to compress all the information needed in path planning. With the help of that the efficiency of the fuzzy logic controller is improved significantly which is also employed to calculate the speed of the robot. Benefit from the fuzzy logic controller and the dangerous region predicted technique, our method is more robust in dynamic environment with uncertainty. And the results of both experiments and simulations have proved that our method get a better effect than the conventional ones.