Abstract: Due to its simplicity and high efficiency,the artificial potential field method has been widely focused and used for autonomous robots.At present,potential field methods have received many accolades in dealing with path planning in static environments or dynamic uniform environments,but unnecessary obstacle avoidance or collision caused by excessive relative velocity of the mobile robot with respect to obstacle in a non-uniform environment will appear .Thus,this paper defined a new potential function with the relative displacement.The relative velocity and the relative acceleration factors are incorporated into the attractive potential function.The obstacle avoidance prediction and the velocity-decreasing collision avoidance strategy are brought into the repulsion potential function,which make the robot can not only track target with variable motion,but also keep the same movement trend with target and remove largely the unnecessary obstacle avoidance.When the relative velocity of the mobile robot with respect to obstacle is larger,the robot can also take an obstacle avoidance measure in advance.The simulation result verifies the effectiveness of method proposed.

Key words: Path planning,Dynamic obstacle avoidance,Mobile robot