Generating dynamically stable walking patterns for humanoid robots using quadratic system model

The approximation of humanoid robot by an inverted pendulum is one of the most used model to generate a stable walking motion. Many studies have been carried out to improve the reliability of this model. One of the proposed models is the quadratic system model, which has been validated by conducting real experiments on the humanoid robot. In this paper, we propose several controlling algorithms for the quadratic system. Some of these algorithms are devoted for on-line and off-line walking pattern generation algorithms for the humanoid robot. Dynamically stable walking patterns have been generated in order to validate these algorithms. The stability and feasibility of walking patterns have been confirmed using dynamical simulation and conducting real experiments on the humanoid robot HRP-4C.