Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

MOBIUS: A Multi-Modal Bipedal Robot that can Walk, Crawl, Climb, and Roll

Alexander Schperberg, Yusuke Tanaka, Stefano Di Cairano, Dennis Hong

TL;DR

MOBIUS introduces a cohesive multi-modal bipedal robot capable of walking, crawling, climbing, and rolling within a single morphology. The authors fuse model-free reinforcement learning for planning/control with model-based MPC baselines, enhanced by an admittance force controller and a Reference Governor for safety, and orchestrate mode choice via a MIQCP high-level planner. Hardware design emphasizes high-DoF arms with GOAT grippers and flat-footed legs, plus back rails for rolling and pinching-grasp climbing, enabling dynamic transitions and full-body load support. The work demonstrates sim-to-real transfer, robust standup and transition maneuvers, vertical climbing with pinch grasp, and energy-aware multimodal planning, highlighting the importance of closely coupling morphology, planning, and control for versatile mobile loco-manipulation.

Abstract

This article presents a Multi-Modal Bipedal Intelligent Urban Scout robot (MOBIUS) capable of walking, crawling, climbing, and rolling. MOBIUS features four limbs--two 6-DoF arms with two-finger grippers for manipulation and climbing, and two 4-DoF legs for locomotion--enabling smooth transitions across diverse terrains without reconfiguration. A hybrid control architecture combines reinforcement learning-based locomotion with model-based predictive and admittance control enhanced for safety by a Reference Governor toward compliant contact interactions. A high-level MIQCP planner autonomously selects locomotion modes to balance stability and energy efficiency. Hardware experiments demonstrate robust gait transitions, dynamic climbing, and full-body load support via pinch grasp. Overall, MOBIUS demonstrates the importance of tight integration between morphology, high-level planning, and control to enable mobile loco-manipulation and grasping, substantially expanding its interaction capabilities, workspace, and traversability.

MOBIUS: A Multi-Modal Bipedal Robot that can Walk, Crawl, Climb, and Roll

TL;DR

MOBIUS introduces a cohesive multi-modal bipedal robot capable of walking, crawling, climbing, and rolling within a single morphology. The authors fuse model-free reinforcement learning for planning/control with model-based MPC baselines, enhanced by an admittance force controller and a Reference Governor for safety, and orchestrate mode choice via a MIQCP high-level planner. Hardware design emphasizes high-DoF arms with GOAT grippers and flat-footed legs, plus back rails for rolling and pinching-grasp climbing, enabling dynamic transitions and full-body load support. The work demonstrates sim-to-real transfer, robust standup and transition maneuvers, vertical climbing with pinch grasp, and energy-aware multimodal planning, highlighting the importance of closely coupling morphology, planning, and control for versatile mobile loco-manipulation.

Abstract

This article presents a Multi-Modal Bipedal Intelligent Urban Scout robot (MOBIUS) capable of walking, crawling, climbing, and rolling. MOBIUS features four limbs--two 6-DoF arms with two-finger grippers for manipulation and climbing, and two 4-DoF legs for locomotion--enabling smooth transitions across diverse terrains without reconfiguration. A hybrid control architecture combines reinforcement learning-based locomotion with model-based predictive and admittance control enhanced for safety by a Reference Governor toward compliant contact interactions. A high-level MIQCP planner autonomously selects locomotion modes to balance stability and energy efficiency. Hardware experiments demonstrate robust gait transitions, dynamic climbing, and full-body load support via pinch grasp. Overall, MOBIUS demonstrates the importance of tight integration between morphology, high-level planning, and control to enable mobile loco-manipulation and grasping, substantially expanding its interaction capabilities, workspace, and traversability.

Paper Structure

This paper contains 55 sections, 21 equations, 20 figures, 9 tables, 3 algorithms.

Table of Contents

  1. Introduction
  2. Contributions
  3. Paper Organization
  4. Related Work
  5. Hardware for Multi-Modality
  6. Multi-Modal Planning and Control
  7. Hardware and System Mechanisms
  8. Overall Design Challenges
  9. Multi-Modal Requirements
  10. Multi-Modalities
  11. Bipedal Locomotion
  12. Crawling Locomotion
  13. Rolling Locomotion
  14. Vertical Mobility: Climbing and Pull-Up
  15. Transitions
  16. ...and 40 more sections

Figures (20)

  • Figure 1: MOBIUS: Multi-modal Operations Bipedal Intelligent Urban Scout.
  • Figure 2: MOBIUS overall structure rendering.
  • Figure 3: Kinematic ranges of the MOBIUS limb modules.
  • Figure 4: MOBIUS range diagram. Each mode has various ranges of motion speeds and reachable heights.
  • Figure 5: Fatigue S-N curve for $1045$ carbon steel approximated from mocko2014influence and the estimated gear teeth strength.
  • ...and 15 more figures