A compound control method based on the adaptive neural network and sliding mode control for inertial stable platform

Abstract

To improve the control performance of the inertial stable platform (ISP), a compound control method based on the adaptive neural network and sliding mode control method is proposed to deal with the system model uncertainties and the disturbances. A sliding mode control is designed for the ISP system to achieve the initial stability. Moreover, an adaptive neural network is proposed to approximate the system uncertainties and unknown disturbances to argument the control performance. Based on the current state error information, the weight matrix of adaptive neural network can be updated on line. Therefore, the adaptive neural network can be constructed directly without priori training. The applicability of the proposed method is validated by a series of simulations and flight tests. The control method can improve the attitude stabilization accuracy of payloads effectively during the flight process.