Neural networks adaptive control for bilateral teleoperation system with uncertain dynamics and kinematics

This paper is aimed at the uncertain dynamics and kinematics of the master-slave manipulators for nonlinear bilateral teleoperation system in the presence of constant time delays and external disturbances in the task space. On the basis of neural networks adaptive control methods, we propose an integral-type neural networks adaptive sliding-mode controller to solve the problem of dynamic uncertainties. Moreover, an adaptive approach about Jacobian matrix deals with the estimate of the unknown kinematic parameters for the manipulators' endeffectors. By applying the integral-type Lyapunov functions, it proves that the position and velocity tracking errors between the master and the slave tend to zero asymptotically. Finally, the effectiveness of the proposed approaches is shown by performing simulations.