Walking is a complex task influenced by many factors. It is still not well understood how single parts as mechanics, sensor feedback and according control components are integrated to the very robust and adaptive task ‘walking’. So one possible way to address this task is to look at single components, as passive mechanics, and analyze their abilities for walking [1,2]. Or rhythmic movement mechanisms in vertebrates are analyzed, like central pattern generator mechanisms, to produce muscle activation patterns [3]. The combination of mechanics and rhythmic actuation leads to more robust walkers [4]. In the following a new biomechanical model for stepping movements in the frontal plain is introduced. This model bases on passive dynamics actuated by a neural oscillator network. It concentrates on low level generation of basic movement patterns which allow different stepping tasks as stepping in place, stepping up and down or stepping to the side. For the case stepping in place, it is shown that there exist periodic movements which are stable in the sense of a limit cycle, though the movement is varied in frequency and amplitude.

This content is only available via PDF.
You do not currently have access to this content.