Neural-network based adaptive sliding mode control for Takagi-Sugeno fuzzy systems

In the present study, the adaptive sliding mode control (ASMC) strategy is investigated for a class of complex nonlinear systems with matched and unknown nonlinearities and external disturbances. The nonlinearities and external disturbances are approached by a Gaussian radial basic neural network. A Takagi-Sugeno (T-S) fuzzy model based integral switching function is introduced to solve the ASMC problem, which eliminates the constrain that input gains required to share a common matrix in all fuzzy rules. Then, the switching control term is represented as a proportional integral (PI) control format to reduce the chattering phenomenon. Based on the Lyapunov theory, a set of existence conditions of the sliding mode controller are given such that the stability of the control systems can be guaranteed. Finally, a experimental simulation is utilized to verify the effectiveness of the proposed sliding mode control (SMC) strategy.

Automated summary

In this study, the adaptive sliding mode control (ASMC) strategy is investigated for complex nonlinear systems with matched and unknown nonlinearities and external disturbances. A Gaussian radial basic neural network is used to approximate the nonlinearities and external disturbances. A Takagi-Sugeno (T-S) fuzzy model based integral switching function is introduced to solve the ASMC problem and eliminate the constraint on input gains. The switching control term is represented as a proportional integral (PI) control format to reduce chattering phenomenon and the Lyapunov theory is used to guarantee the stability of the control systems. An experimental simulation is conducted to verify the effectiveness of the proposed sliding mode control (SMC) strategy.