RBF Neural Network sliding mode Control of Onboard Craning Manipulator Based on Backstepping

A dynamic model based on Lagrange principle is derived for an onboard craning manipulator system while the impacts of both gravity and coupling effect are simultaneously considered in this paper. Combining with the designed dynamic model, a trajectory tracking control method is proposed through the approximation theory of neural network system, because the boundary of parameters uncertainties and external disturbances could not be attained easily in the actual system. To overcome the impact of unknown and bounded uncertainty, a backstepping RBF neural network sliding mode controller is designed. Finally these simulation results show that the proposed control scheme can realize the trajectory tracking performance of system. And both the tracking errors and approximation errors can asymptotically converge to zero.