### Meeting Abstract

**S1-3.12** Jan. 5 **Effects of Leg Number and Posture for a Class of Simple Three-dimensional Point-mass Running Models** *SEIPEL, J.*; FULL, R.J.; HOLMES, P.; Univ. of California, Berkeley; Univ. of California, Berkeley; Princeton Univ.* jseipel@princeton.edu

Many legged animals produce similar net ground reaction forces during steady running - like an idealized point-mass pogo-stick. The simple pogo-stick analogy reduces the forces along a **steady**-gait to a single effective springy leg regardless of the system's complexities. Yet, morphology matters to the dynamics and energetics of such systems, even for a class of highly idealized zero-friction point-mass models. Morphological details of form, such as leg number and posture, make significant 'first order' contributions to the **dynamics** of running gaits, and thus to the stability of perturbations from steady-state periodic running. A phenomenological, single effective leg cannot reproduce these effects. Furthermore, while a simple passive monopod model can complete the task of running with no energy expenditure, the inclusion of sprawled postured legs requires actuation and thus energy expenditure significantly higher than what would seem necessary for an otherwise simple running gait. We present a laterally sprawled kangaroo model with bilateral symmetry and a tripod-sprawled cockroach model. We compared these animal models with the alternating tripod gait of the robotic hexapod RHex. We found that a sprawled posture can benefit dynamic stability. In particular, gaits of the cockroach-like model were quite robust even though they consumed energy. This result, along with work by Kubow and Full (1999) and Seipel et al. (2004) which modeled the tripod of the cockroach in the horizontal plane and where the body could yaw, supported the hypothesis that dynamic stability is an important factor governing cockroach morphology and gait selection, despite the consequence of higher energy expenditure. NSF FIBR grant and NSF Fellowship.