摘要：For the carrying lower extreme exoskeleton system is non-linear and have variable parameters, which is usually subject to external dusturbances. The virtual torque control method is a well-established carrying system control technique, which needs no sensors between the pilot and the human-machine interface, instead, the controller estimates, based on measurements from the exoskeleton suits only, how to move so the pilot feels very little force. However, its main disadvantage lies on solveing the inverse dynamic model that is not easily obtained, which also need an exactly known dynamic model that is not realizable in practice. To overcome this issue, we propose building a virtual prototype model of the lower extremetry exoskeleton in SimMechanics. Using torque input and joint angle output data of the the virtual prototype model, the wavelet neural network is trained to building the inverse dynamics model which take the joint angle data as input and the torque data as output. In cases where some of the parameters of the plant undergo a change or the system is disturbed, poor performance may result. To cope with this drawback, based on Lyapunov stability principle, the rapid terminal sliding-mode robust adaptive controller is designed to control the serious nonlinear lower extreme exoskeleton system, and the convergence performance of the controller is analysed. Theoretical analyse and simulation results test the feasibility and validity of this control method.
