Painlevé's paradox and dynamic jamming in simple models of passive dynamic walking

Research output: Contribution to journalArticlepeer-review

Abstract

Painlevé's paradox occurs in the rigid-body dynamics of mechanical systems with frictional contacts at configurations where the instantaneous solution is either indeterminate or inconsistent. Dynamic jamming is a scenario where the solution starts with consistent slippage and then converges in finite time to a configuration of inconsistency, while the contact force grows unbounded. The goal of this paper is to demonstrate that these two phenomena are also relevant to the field of robotic walking, and can occur in two classical theoretical models of passive dynamic walking - the rimless wheel and the compass biped. These models typically assume sticking contact and ignore the possibility of foot slippage, an assumption which requires sufficiently large ground friction. Nevertheless, even for large friction, a perturbation that involves foot slippage can be kinematically enforced due to external forces, vibrations, or loose gravel on the surface. In this work, the rimless wheel and compass biped models are revisited, and it is shown that the periodic solutions under sticking contact can suffer from both Painlevé's paradox and dynamic jamming when given a perturbation of foot slippage. Thus, avoidance of these phenomena and analysis of orbital stability with respect to perturbations that include slippage are of crucial importance for robotic legged locomotion.

Original languageEnglish
Pages (from-to)64-80
Number of pages17
JournalRegular and Chaotic Dynamics
Volume19
Issue number1
DOIs
StatePublished - Feb 2014

Keywords

  • Painlevé paradox
  • dry friction
  • multibody dynamics
  • passive dynamic walking
  • rigid body contact

All Science Journal Classification (ASJC) codes

  • Mathematics (miscellaneous)

Fingerprint

Dive into the research topics of 'Painlevé's paradox and dynamic jamming in simple models of passive dynamic walking'. Together they form a unique fingerprint.

Cite this