An Experimental Study of Three-Dimensional Passive Dynamic Walking with Flat Feet and Ankle Springs

Passive dynamic bipeds were first studied by McGeer (McGeer, 1990) as inspired by a bipedal toy described in (McMahon, 1984). Passive dynamic walkers can walk down a shallow slope without actuators and controllers(McGeer, 1990; 1993). McGeer has built passive walkers that exhibit steady motion using a Poincaré map, which he called as a stride function, to analyze the gaits(McGeer, 1990; 1993). This method is quite useful and is independent of the biped model. The key idea that he examined is the stability of the entire step-to-step motion, and not the local stability at every instance. This analysis is also rather useful for actuated bipeds (Hobbelen and Wisse, 2007). Firstly, McGeer studied two elementary passive walkingmodels derived from awagonwheel. One model was a rimless wheel model on a slope, and the other a synthetic wheel model on level ground as shown in Fig. 1. The motion of the models is constrained to the sagittal plane. Each model captures the fundamental mechanism of passive dynamic walking.