To solve the path planning problem of mobile robots in complicated environments, combining the small computational amount of artificial potential field and the adaptive regulation ability of artificial immune network, an improved immunity path planning algorithm is presented. To further improve the search ability and convergence of immune network, the planning results of artificial potential field are taken as the prior knowledge to construct the guidance weight, and the antibody instruction definition and distance changes after the antibody transition are taken as the variables to construct the antibody transmission probability operator. Compared with the correlative planning algorithms, simulation results indicate that the optimal planning ability and network convergence of proposed algorithm is highly improved.