In order to solve the problem that the parameters in the core module of the auto  disturbance  rejection controller are difficult to be set, this paper presents a design method of the extended state observer using the improved particle swarm optimization algorithm. In this paper, the mathematical model of the extended state observer is studied, and the factors influencing the tracking accuracy of the observed signal are analyzed. The concrete steps of the parameter design method are given. The design method chooses the appropriate fitness function as the performance optimization index, increases the selfadaptability of inertia factor, and introduces the speed clamp to prevent the local optimum. Through the theoretical calculation and simulation analysis, the results verify the correctness and the strong robustness of the designed method. With this method, the parameters of the other modules of the autodisturbance rejection controller can be determined.