Abstract: The artificial potential field method, as a path planning algorithm, is used for the path planning of clumsy mobile robots, agents and so on because it can provide a smooth route. Due to the inevitable existence of local minimum and discretization of algorithms during algorithm execution, the calculation of the route point may fall into the local minimum trap, which may result in the infinite loop of algorithmic program and the oscillation of path point. For local minimum trap, an improved virtual obstacle method is proposed to overcome this problem. The definition of threat area is introduced to determine the location of virtual obstacle, and a determining standard is put forward. A filtering oscillation point method is proposed to solve the problem of path point oscillation. The simulated results show that the route points which are stuck in local minimum trap can escape from the trap successfully using the improved virtual obstacle method. Additionally, the oscillation route points can be effectively eliminated to obtain a relatively smooth route by the filtering oscillation point method. Key

Key words: artificialpotentialfield, localminimum, virtualobstacle, filteringoscillationpoint, smoothpath