Abstract: The artificial potential field method is a simple and efficient path planning algorithm for mobile robots.Aiming at a kind of goal unreachable problem in traditional artificial potential field methods,an improved algorithm which changes the angle of repulsion at the local minimum point and sets the virtual local minimum area was proposed for problem-solving.The genetic algorithm was also introduced to optimize the parameters,i.e.the revolved angle of repulsion and the radius of the virtual local minimum area for the improved artificial potential field algorithm.It is proved that the proposed algorithm can plan out a simple,smooth and safe optimum path connecting the start point and the end point by simulation experiments.