Skip to main navigation Skip to search Skip to main content

A Unified Foot-Terrain Interaction Model for Legged Robots Contacting With Diverse Terrains

  • Harbin Institute of Technology
  • Automotive Engineering College
  • Toronto Metropolitan University

Research output: Contribution to journalArticlepeer-review

Abstract

Interaction with the terrain is essential for legged robots to adapt to complex environment. Field terrain could be steep, slippery, and muddy. Legged robots may sink on sand or mud and slip on ice or snow. To address such problems, foot-terrain interaction models have been developed to preestimate the sink or slip state of the robot and models are switched to adapt to different terrain. However, it is difficult to switch the model precisely, which causes problems in its application to complex terrain in the field. This article proposes a unified foot-terrain interaction model to avoid model switching in multiphysical characteristic terrains. Specifically, a normal foot-terrain interaction model is formulated to characterize the dynamic sinkage of robot foot into soft terrain which combines velocity and loading effects on the sinkage exponent. Furthermore, a sinkage-slip model is proposed to reflect sliding friction and bulldozing resistance. Finally, by generalizing the models between different terrains, a unified model of hard-soft-slippery terrain is completed in both normal and tangential directions. Single-foot and robot-movement experimental results show that the proposed model can adapt to different field terrains with high accuracy and efficiency.

Original languageEnglish
Pages (from-to)2661-2672
Number of pages12
JournalIEEE/ASME Transactions on Mechatronics
Volume29
Issue number4
DOIs
StatePublished - 2024

Keywords

  • Dynamic contact
  • foota-terrain interaction model
  • legged robots
  • multiphysical characteristic terrain

Fingerprint

Dive into the research topics of 'A Unified Foot-Terrain Interaction Model for Legged Robots Contacting With Diverse Terrains'. Together they form a unique fingerprint.

Cite this