RBF adaptive sliding control for five-axis flexible satellite

With consideration of the flexible attachment effect, a robust sliding mode control method is proposed to realize RBF optimization of 3-DOF satellite with 2-DOF load. RBF neural network´s approximation characteristic enables us to estimate the nonlinear function of satellite dynamics equation, furthermore, to adjust the sliding mode control law. Meanwhile, adaptive law is adopted to optimize the weights of neural network, so that satellite attitude and load attitude reach the expectation. Simulation results reveal that, this method far surpass traditional control method in rapidity, robustness and accuracy. Moreover, it´s timeliness values much during practical application.