Adaptive synchronous artificial neural network based PI-type sliding mode control on two robot manipulators

Abstract

An adaptive synchronous proportional-integral (PI)-type sliding mode control is developed for two cooperative robot manipulators handling a lightweight beam. This approach is under implicit communication between robots in which each robot manipulator does not need to have any information about the other. A class of sliding mode control which is insensitive and robust in the presence of the uncertainties and external disturbances with no chattering is applied. In the sliding mode control investigating PI sliding surface guarantee the asymptotic stability in compare with the classic sliding mode control. A feed forward neural network is applied to compensate dynamic model uncertainty. The adaptive synchronization method is presented to solve the parameter uncertainty in the trajectory of each robot manipulator with respect to the reference to handle the object accurately and smoothly in the desired trajectory. The stability analysis of the proposed scheme is guaranteed by Lyapunov method. In the simulation results, the convergence of trajectory tracking error and synchronization error to zero is reveal the performance of proposed scheme.