A novel adaptive neural network integral sliding-mode control of a biped robot using bat algorithm

Abstract

This paper proposes a new adaptive neural network integral sliding-mode controller using a bat algorithm (BA-ANNISMC) to control a biped robot. The conventional integral sliding-mode controller (ISMC) is discontinuous in nature due to the combination of nominal control and discontinuous feedback control. The phenomenon of chattering occurs when there is a discontinuity in feedback control. An adaptive neural network is applied to estimate the unknown disturbances to the system. Therefore, by using an adaptive neural network, the chattering phenomena will be eliminated. The proposed controller parameters are tuned using a bat algorithm. The stability of the adaptive neural network integral sliding-mode controller (ANNISMC) is proved by Lyapunov theory. In order to show the effectiveness of the proposed controller, its performance is compared with three other controllers such as a conventional sliding-mode controller (SMC), ISMC and ANNISMC. The results of the numerical simulation clearly indicate the effectiveness of the BA-ANNISMC controller when considering chattering reduction.