Towards Terradynamics of Dynamic Legged Locomotion in Complex 3-D Terrains

Abstract

Over the last few years, the emerging field of terradynamics, analogous to aero- and hydrodynamics for flying and swimming, has begun to advance our understanding of how movement in the real world emerges from effective locomotor-terrain interaction. The majority of terradynamic studies so far have focused on locomotion on/within relatively uniform granular media like sand, and the first terradynamic models from these studies have advanced our understanding of the locomotion of desert-dwelling animals and led to robots capable of traversing sandy environments. However, many natural and artificial terrains (such as dense vegetation and building rubble) are beyond relatively uniform granular media and have complex three dimensional geometry, configuration, and mechanical properties. Therefore, we need new terradynamics to describe the physical interaction and allow prediction of movement in these types of complex 3-D terrains. We propose to advance terradynamics for legged locomotion in complex 3-D terrains. Specifically, we will test the hypothesis that vibrations and perturbations from locomotor-terrain interaction during legged locomotion serve as Òenergy fluctuationsÓ to allow the formation of locomotor pathways. To test our hypothesis, we will address four research aims: (1) Creation of new terrain devices to partition and parameterize complex 3-D terrains. (2) Animal experiments to measure locomotor performance and pathways and test the hypothesis. (3) Robophysics experiments to vary locomotor and terrain parameters to discover general principles and test the hypothesis. (4) Physics modeling to understand formation of locomotor pathways. This fundamental research will make a practical impact by providing a largely missing cornerstone for mobile robotics. Most current mobile robots still navigate the world primarily by avoiding obstacles using geometry-based environmental models. However, to ensure successful traversal through complex 3-D terrains, during which physical contact is essential, we must also understand the mechanics of locomotor-environment interaction. The new terradynamics emerging from this research will provide mechanics-based environmental models and enable legged robots to begin to traverse, not merely avoid, a diversity of novel environments.

Document Details

Document Type
DoD Grant Award
Publication Date
Oct 16, 2018
Source ID
W911NF1710346

Entities

People

  • Chen Li

Organizations

  • Army Contracting Command
  • Johns Hopkins University
  • United States Army

Tags

Readers

  • Computational Fluid Dynamics (CFD)
  • Robotics and Automation.

Technology Areas

  • AI & ML
  • AI & ML - Autonomous Systems
  • Autonomy