Home Research BCCN I Project Details D4 - Towards a hybrid active/passive bipedal walking prosthesis
Personal tools

D4 - Towards a hybrid active/passive bipedal walking prosthesis

Michael Herrmann, Lüder Mosler, and Walter Paulus

Walking kinematics has been well studied in normals (Troje, 2002) and in walking disorders (Whittle, 1996). In parallel, bipedal robots with computationally controlled humanoid gait patterns were realized more than ten years ago (Raibert, 1991), when the computationally complex control problems had found a preliminary solution, and have continuously improved since. Present technical realization in robots still suffers from the mismatch of the desired gaits and the intrinsic dynamics of the hardware. The idea of passive dynamic walking (McGeer, 1993) requires fine-tuned hardware properties and herewith allows for smooth and efficient bipedal gaits. Only recently passive dynamics and and active control approaches have been combined in order to optimize both efficiency, energy consumption, and flexibility (Ohta 1999). The applicants have recently (Mayer 2004) studied asymmetry in gait patterns as well as in the mechanical parts of an artificial walker. Some 25 contributions have been published on visual stabilization of posture (overview in Paulus 1993).

Belongs to Group(s):