The invention discloses an auto-disturbance rejection position servo system optimization design method based on an improved CPSO. By aiming at problems of permanent magnet synchronous motor servo systems on high position control precision, fast response, and stable performance, a double-loop control structure is adopted, and a PMSM auto-disturbance rejection position servo control system is established. By aiming at a parameter setting problem of an auto-disturbance rejection position controller, the improved Chaos Particle Swarm Optimization (CPSO) is provided. By adopting the CPSO, a position of a particle is initialized according to cubic chaotic mapping, and an index self-adaptive way having adjustable parameters is used to adjust inertia weight in a non-linear way, and at the same time, the position of the particle is updated by adopting a chaos and stability alternate way, and therefore the convergence rate and the global optimization ability of the CPSO are effectively improved, and the CPSO is used for the optimization of the auto-disturbance rejection position controller parameters. By combining with a fitness function including position control requirements, the optimization design of the PMSM position servo control system is realized, the position control precision and the response speed of the servo system are improved, and a strong disturbance rejection ability is provided.