Due to the inappropriate value of inertia weight and the diversity of particle population decreases in late iteration, traditional particle algorithm in the process of mobile robot path planning easy falls into the local optimal solution problem. Aiming at this problem, a path planning method of mobile robot with improved particle swarm optimization is proposed. Firstly, a grid map model of robot path planning is established. On this basis, the traditional particle swarm optimization algorithm is improved. The weight dynamic adjustment method based on the concept of similarity is introduced to update the particle update rate with the various stages of the optimization process, by introducing the immune information regulation mechanism in the immune algorithm to increase the diversity of particles to enhance its ability to get rid of the local optimum. The simulation results show that the proposed method can obviously improve the searching ability of the best path and its comprehensive performance is better than the traditional particle swarm optimization.