Skip to Main Content
Under the framework of hybrid control, RBF neural network is used to compensate for all the uncertainties from robot dynamics and unknown environment at first. It can improve the capability of the adaptive to environment stiffness when the end-effector contacts with the environment. It does not require any a prior knowledge on the upper bound of system uncertainties. Moreover, we use GASA algorithm to find the optimal structure weight of RBF neural network. Simulation results have shown better force/position tracking when neural network is used.