Neural network adaptive control of a Stewart mechanism-based active vibration isolation platform

For a Stewart mechanism-based active vibration isolation platform,a multiple input-multiple output adaptive vibration isolation controller based on radial basis function network(RBFN) is proposed.Considering vibration effects on the vibration isolation platform,a dynamic model of the platform in the workspace is developed.An online adaptive tuning rule for updating weights,centers and widths of the RBFN is derived to approximate the nonlinear dynamics of the system.By using the Lyapunov synthesis approach,it is proved that the filtered error of the closed-loop system and the parameters of the RBFN are uniformly ultimately bounded in the presence of bounded disturbance force and bounded neural network approximation error.Simulation results show that the developed RBFN controller can effectively attenuates low frequency vibrations in all six degress of freedom.