RBF Neural Network Sliding Mode control for Aeronautical Remote Sensing Stable Platform

The Aviation Remote Sensing Stabilization Platform (ARSSP) is widely used in spacecraft navigation, positioning, aerial photography and other tasks. In the process of aerial photography and other tasks, the existence of nonlinear interference and model uncertainties and other factors cause the accuracy of ARSSP's boresight pointing accuracy and stability to deteriorate. Therefore, this paper uses an RBF neural network sliding mode control (RBF-NNSMC) algorithm to optimize the ARSSP controller. Firstly, the gear drive system of ARSSP is designed according to the function of ARSSP, and the system dynamic model is established. Then, according to the dynamic model of the system, the RBF-NNSMC algorithm is adopted to reduce the nonlinear interference and approach the uncertain items of the model, the weight of RBF neural network is adjusted by the minimum parameter learning method. Stribeck friction interference model is introduced as the main interference term, in which the saturation function is employed to reduce the chattering, and its stability is proved based on the Lyapunov theory. The simulation experiment shows that the boresight pointing error of the ARSSP system can reach 0.002rad, and the system response time is 0.1s. Therefore, this method can effectively reduce the interference of nonlinear factors to the system. At the same time, the boresight pointing accuracy of ARSSP is significantly improved, which enhances the robustness and stability of the system.