Abstract: A path planning method based on both the benefits of global and local path planners and those of deliberate and reactive behavior planners is proposed for indoor mobile robot path planning in dynamic and unstructured environments.The global path planner uses A^* algorithm to generate a series of sub-goal nodes to the target node,and the local path planner adopts an improved potential field method to smooth and optimize the path between the adjacent sub-goal nodes.Taking into full consideration the kinematical constraints of the mobile robot,this method can not only effectively generate a global optimal path using the known information,but also handle the stochastic obstacle information in time.Results of simulations and realistic running in complex indoor environments validate the effectiveness of the presented algorithm.