A fuzzy CMAC compensator for sliding mode control

A self-organizing neuro-fuzzy compensator which cooperates with a sliding mode controller is proposed to tackle physical learning control problems in this paper. The sliding mode, hyperplane or a desired controller is used to drive the state variables to a specific switching trajectory. A fuzzy cerebellar model articulation controller (FCMAC) with self-organizing capability functions as a forward compensator to reduce the chattering effect or cancel the influence of system uncertainties. The method is actually based on the concept of the conventional CMAC and reinforcement learning. Simulations on balancing a cart-pole system are given to illustrate the performance and applicability of the proposed control scheme.