Intelligent Robust Adaptive Trajectory and Force Tracking Controller for Holonomic Constrained Nonholonomic Mobile Manipulators

The aim of this paper is to design a robust adaptive neural network (NN) based trajectory/force control scheme for holonomic constrained nonholonomic mobile manipulators in the presence of uncertainties such as load variation from task to task and external disturbances etc. The feedforward neural network is employed to learn a highly nonlinear function, which requires no preliminary learning. The control purpose is not only to achieve the desired trajectory but also to ensure that the constraint force converge to the desired force. Moreover, an adaptive compensator is developed to eliminate the effect of the disturbance term of neural network approximation error and un-modeled dynamics. A key feature of this compensator is that the prior information of the disturbance bound is not required. The stability of the closed error system is proved using Lyapunov stability analysis. Finally, a comparative simulation study with a model based robust control scheme is presented.