Meeting Abstract

P2-156   -   Questions around the catapult mechanism in human legged locomotion Kiss, B*; Buchmann, A; Renjewski, D; Badri-Sprowitz, A; Max Planck Institute for Intelligent Systems; Technical University of Munich; Technical University of Munich; Max Planck Institute for Intelligent Systems kiss@is.mpg.de http://dlg.is.mpg.de

Our aim is to describe the components and the function of the catapult mechanism in human legs, and replicate it on a robotic model. Hof (1983) was the first to propose that the ankle power burst during the late stance phase of human gait is preceded by a slower energy storage phase which makes the human leg comparable to a catapult. Different types of catapult mechanisms were observed in other animals as well, such as horses (Wilson et al., 2003), locusts, frogs, fleas, click beetles (Alexander & Bennett-Clark, 1977) and fish (Aerts, 1987). A catapult has three main components: an elastic component, a block, and a catch with or without escapement. The catch has been thoroughly studied in, for instance, locusts and frogs but has not been fully identified in the human leg yet. The release of the catch can be triggered by the imbalance of the internal and external moments around a certain body part or joint, (Wilson et al., 2003) or by an active neural command (Aerts, 1997). The catapult’s function in human locomotion is discussed controversially, i.e., what happens to the released energy. During touch-down of the leading leg collision losses arise which must be restored. Push-off work of the trailing leg could redirect the center of mass velocity and considerably reduce the effect of the collision if the knee joint’s buckling just before push-off is not taken into account (Zelik et al., 2014). However, a flexed knee joint is assumed to transfer only a fraction of the push-off energy into the trunk, and with that in mind, push-off work could power leg swing instead (Lipfert et al., 2014). Deeper understanding of the catapult mechanism bears great potential for improving orthoses, prostheses, legged robots, and could be leveraged in gait rehabilitation. We will present our first insights from implementing a robotic model in hardware and simulation.