Partial Human Data in Design of Human-Like Walking Control in Humanoid Robotics

In this paper a human-based control strategy is proposed for walking of humanoid robots. Its basic idea lies in the control of the CoM as well as the relative position of robot's feet. Through the proposed study, it is shown that this reduced set of data allows to fully describe and reproduce a whole body human-like walking. In addition of reducing the number of controlled variables, the proposed control strategy has the advantage of not requiring a complete walking cycle decomposition due to its continuous character. Simulations results are presented to show the effectiveness of the proposed control scheme.