Neuro-fuzzy ZMP control of a biped robot

Abstract

This paper describes the control of a biped robot that possesses innovative characteristics, namely regarding its autonomy, becoming important the existence of a wireless transmission that binds software running in a PC to the robot in an efficient way and without physical links. It is used a radio transmission link and a power battery is carried by the robot. In this work a 1st order Takagi-Sugeno-Kang type neuro fuzzy net is proposed, based on experimental and simulation data. This net uses the ZMP position and its variation as input and the longitudinal correction of the robot's body is obtained as the output. The ZMP is obtained making use of the force distribution on four force sensors placed under each robot foot.