Optimization based exploitation of the ankle elasticity of HRP-2 for overstepping large obstacles

This paper proposes a new generic strategy to investigate the dynamic limits of the humanoid robot HRP-2 based on whole body optimal control optimization. In this study we exploit the intuitive access to complex motion characteristics, given by optimal control, to effectively resolve a major technical coupling effect, namely between the ankle elasticity and the stabilizing algorithms. Control efforts are reduced to get a clearer view of the actual system limits and to exploit its capacities at maximum. As showcase we decided to focus on a stepping motion over a cylindrical obstacle. This study is further supported by real experiments on the HRP-2 14 robotic platform and the present maximum of a dynamically overstepped obstacle was extended to 20cm (height) × 11cm (width) (including safety margin) without multi-contact support.