A parameter setting method is proposed for the nonlinear extended state observer in the Active Disturbance-Rejection Controller (ADRC) by using Back-Propagation Neural Network (BPNN). This method simplifies the complicated parameter setting process of the nonlinear extended state observer. Simulation is made to the BPNN based ADRC system. The simulation results show that this method can significantly enhance the disturbance isolation of the airborne optoelectronic stabilized platform, and the control effect is obviously superior to that of the traditional active disturbance-rejection control method, which is of great significance for improving the optical-axis stabilizing accuracy of the airborne optoelectronic stabilized platform.