이족 로봇의 안정한 걸음새를 위한 자기 회귀 웨이블릿 신경 회로망을 이용한 적응 백스테핑 제어

Adaptive Backstepping Control Using Self Recurrent Wavelet Neural Network for Stable Walking of the Biped Robots

This paper presents the robust control method using a self recurrent wavelet neural network (SRWNN) via adaptive backstepping design technique for stable walking of biped robots with unknown model uncertainties. The SRWNN, which has the properties such as fast convergence and simple structure, is used as the uncertainty observer of the biped robots. The adaptation laws for weights of the SRWNN and reconstruction error compensator are induced from the Lyapunov stability theorem, which are used for on-line controlling biped robots. Computer simulations of a five-link biped robot with unknown model uncertainties verify the validity of the proposed control system.