Adaptive multi-model CMAC-based supervisory control for uncertain MIMO systems

In this paper, an adaptive multi-model CMAC-based controller (AMCBC) in conjunction with a supervisory controller is developed for uncertain nonlinear MIMO systems. AMCBC is a kind of adaptive feedback linearizing controller where nonlinearity terms are approximated with multiple CMAC neural networks. With the help of a supervisory controller, the resulting close-loop system is globally stable. The proposed control system is applied to control a robotic manipulators, where some varying tasks are repeated but information on the load is not defined; it is unknown and varying. It is shown how the proposed controller is effective because of its capability to memorize the control skill for each task using CMAC neural network. Simulation results demonstrate the effectiveness of the proposed control scheme for the robotic manipulators