We show that, near periodic orbits, a class of hybrid models can be reduced to or approximated by smooth continuous-time dynamical systems. Specifically, near an exponentially stable periodic orbit undergoing isolated transitions in a hybrid dynamical system, nearby executions generically contract superexponentially to a constantdimensional subsystem. Under a non-degeneracy condition on the rank deficiency of the associated Poincaré map, the contraction occurs in finite time regardless of the stability properties of the orbit. Hybrid transitions may be removed from the resulting subsystem via a topological quotient that admits a smooth structure to yield an equivalent smooth dynamical system. We demonstrate reduction of a high-dimensional underactuated mechanical model for terrestrial locomotion, assess structural stability of deadbeat controllers for rhythmic locomotion and manipulation, and derive a normal form for the stability basin of a hybrid oscillator. These applications illustrate the utility of our theoretical results for synthesis and analysis of feedback control laws for rhythmic hybrid behavior.
I. INTRODUCTION Rhythmic phenomena are pervasive, appearing in physical situations as diverse as legged locomotion [1], dexterous manipulation [2], gene regulation [3], and electrical power generation [4]. The most natural dynamical models for these systems are piecewise-defined or discontinuous owing to intermittent changes in the mechanical contact state of a locomotor or manipulator, or to rapid switches in protein synthesis or constraint activation in a gene or power network. Such hybrid systems generally exhibit dynamical behaviors that are distinct from those of smooth systems [5]. Restricting our attention to the dynamics near periodic orbits in hybrid dynamical systems, we demonstrate that a class of hybrid models for rhythmic phenomena reduce to classical (smooth) dynamical systems.Although the results of this paper do not depend on the phenomenology of the physical system under investigation, a principal application domain for this work is terrestrial locomotion. Numerous architectures have been proposed to explain how animals control their limbs; for steady-state locomotion, most posit a principle of coordination, synergy, symmetry or synchronization, and there is a surfeit of neurophysiological data to support these hypotheses [6]- [10]. Taken together, the empirical evidence suggests that the large number of degrees-of-freedom (DOF) available to a locomotor can collapse during regular motion to a low-dimensional dynamical attractor (a template) embedded within a higher-dimensional model (an anchor) that respects the locomotor's physiology [1], [11]. We provide a mathematical framework to model this empirically observed dimensionality reduction in the deterministic setting.A stable hybrid periodic orbit provides a natural abstraction for the dynamics of steady-state legged locomotion. This widely-adopted approach has generated models of bipedal [12] . In certain cases, it has been possibl...