In-Place Rotation for Enhancing Snake-like Robot Mobility
Abstract
Gaits engineered for snake-like robots to rotate in-place instrumentally fill a gap in the set of locomotive gaits that have traditionally prioritized translation. This paper designs a Turn-in-Place gait and demonstrates the ability of a shape-centric modeling framework to capture the gait's locomotive properties. Shape modeling for turning involves a time-varying continuous body curve described by a standing wave. Presumed viscous robot-ground frictional interactions lead to body dynamics conditioned on the time-varying shape model. The dynamic equations describing the Turn-in-Place gait are validated by an articulated snake-like robot using a physics-based simulator and a physical robot. The results affirm the shape-centric modeling framework's capacity to model a variety of snake-like robot gaits with fundamentally different body-ground contact patterns. As an applied demonstration, example locomotion scenarios partner the shape-centric Turn-in-Place gait with a Rectilinear gait for maneuvering through constrained environments based on a multi-modal locomotive planning strategy. Unified shape-centric modeling facilitates trajectory planning and tracking for a snake-like robot to successfully negotiate non-trivial obstacle configurations.
Keywords
Cite
@article{arxiv.2203.05043,
title = {In-Place Rotation for Enhancing Snake-like Robot Mobility},
author = {Alexander H. Chang and Patricio A. Vela},
journal= {arXiv preprint arXiv:2203.05043},
year = {2022}
}
Comments
8 pages, 5 figures. Submitted to RA-L (IEEE Robotics and Automation Letters) with IROS 2022 Option