Control of a Nonholonomic Mobile Manipulator Using Neural-Network

In this paper, a control method for a mobile manipulator subjected to kinematic constraint is proposed. It is difficult to control a mobile manipulator because of nonholonomic constraint on the mobile platform and nonlinear dynamics with dynamic interaction between the mobile platform and the manipulator. The neural network is used to estimate mobile manipulator dynamics with the dynamic interaction by on-line learning rule. The stability of the closed-loop system, the convergence of the neural network weight-updating process and boundedness of the neural network weight estimation error are guaranteed by Lyapunov theory.