Adaptive neural network controller of a stewart platform with unknown dynamics for active vibration isolation

To solve the control problem of a Stewart platform with unknown dynamics for multiple degree-of-freedom (DOF) active vibration isolation, an adaptive radial basis function neural network(RBFNN) controller is developed. The RBFNN is employed to approximate the unknown dynamics of the system. And an on-line tuning rule for the parameters of the RBFNN is given based on the e1-modification and gradient algorithms. Meanwhile, a sliding mode control term is incorporated to further improve the robustness of the whole controller against external vibrations. In the presence of bounded vibrations, the uniformly ultimately boundedness of the filter error and the estimation errors of the RBFNN parameters can be guaranteed by the Lyapunov theory. Finally, simulation results demonstrate the proposed controller can effectively attenuate external low frequency vibrations in all six DOF.