ZMP Compensation by On-Line Trajectory Generation for Biped Robots

This paper proposes an adaptive trajectory generation strategy of using on-line ZMP information with an impedance control method. During walking, since the robot experiences various disturbances like pulling or pushing forces, the walking mechanism should have the robustness against those disturbances, and it requires an on-line adaptation ability. If sensed ZMP goes out of the preset ZMP boundary, the on-line trajectory planner generates the trajectory of the base link in the vertical direction to compensate the required moment for recovering stability. The ZMP equation and sensed ZMP information are used in this trajectory generation strategy. In this paper, an impedance controller with impedance modulation is used to control biped robots for stable walking. Computer simulation with a 3-dof environment model for which a combination of a nonlinear and linear compliant contact models is used, shows that the proposed controller performs well. Moreover, the biped robot with the proposed trajectory generator can walk even when it is pushed with a certain amount of force.