Switching feedback controllers based on the maximal CPI sets for stabilization of humanoid robots

Humanoid robots should be able to stand and walk despite reasonable external disturbances. This paper addresses the robustness of a humanoid robot to unknown disturbances. Although there are control methods to absorb the disturbances based on the COG-ZMP inverted pendulum model, they do not consider the physical constraint on ZMP. In this paper, the authors enable the control law to consider the physical constraint explicitly with the maximal CPI set. Furthermore, the switching control framework is applied to a COG-ZMP inverted pendulum model, allowing for an improved robustness to external disturbances. The validity of the proposed method is verified with both of simulation and experiment.