Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Robust Localization, Mapping, and Navigation for Quadruped Robots

Dyuman Aditya, Junning Huang, Nico Bohlinger, Piotr Kicki, Krzysztof Walas, Jan Peters, Matteo Luperto, Davide Tateo

TL;DR

This work tackles robust localization, mapping, and navigation for low-cost quadruped robots using inexpensive RGB-D and IMU sensors in indoor settings. It introduces a modular stack that fuses legged-odometry, contact estimation, scan stabilization, and visual-inertial odometry within a 2D SLAM framework, and integrates these into ROS2-based navigation with RTAB-Map and slam_toolbox. Through extensive ablations and real-world experiments on two platforms, the authors show that scan stabilization is essential for reliable mapping and loop closure, while legged odometry enhances VIO performance and pose consistency. The results demonstrate practical feasibility for autonomous navigation and exploration of cluttered indoor environments with affordable hardware.

Abstract

Quadruped robots are currently a widespread platform for robotics research, thanks to powerful Reinforcement Learning controllers and the availability of cheap and robust commercial platforms. However, to broaden the adoption of the technology in the real world, we require robust navigation stacks relying only on low-cost sensors such as depth cameras. This paper presents a first step towards a robust localization, mapping, and navigation system for low-cost quadruped robots. In pursuit of this objective we combine contact-aided kinematic, visual-inertial odometry, and depth-stabilized vision, enhancing stability and accuracy of the system. Our results in simulation and two different real-world quadruped platforms show that our system can generate an accurate 2D map of the environment, robustly localize itself, and navigate autonomously. Furthermore, we present in-depth ablation studies of the important components of the system and their impact on localization accuracy. Videos, code, and additional experiments can be found on the project website: https://sites.google.com/view/low-cost-quadruped-slam

Robust Localization, Mapping, and Navigation for Quadruped Robots

TL;DR

This work tackles robust localization, mapping, and navigation for low-cost quadruped robots using inexpensive RGB-D and IMU sensors in indoor settings. It introduces a modular stack that fuses legged-odometry, contact estimation, scan stabilization, and visual-inertial odometry within a 2D SLAM framework, and integrates these into ROS2-based navigation with RTAB-Map and slam_toolbox. Through extensive ablations and real-world experiments on two platforms, the authors show that scan stabilization is essential for reliable mapping and loop closure, while legged odometry enhances VIO performance and pose consistency. The results demonstrate practical feasibility for autonomous navigation and exploration of cluttered indoor environments with affordable hardware.

Abstract

Quadruped robots are currently a widespread platform for robotics research, thanks to powerful Reinforcement Learning controllers and the availability of cheap and robust commercial platforms. However, to broaden the adoption of the technology in the real world, we require robust navigation stacks relying only on low-cost sensors such as depth cameras. This paper presents a first step towards a robust localization, mapping, and navigation system for low-cost quadruped robots. In pursuit of this objective we combine contact-aided kinematic, visual-inertial odometry, and depth-stabilized vision, enhancing stability and accuracy of the system. Our results in simulation and two different real-world quadruped platforms show that our system can generate an accurate 2D map of the environment, robustly localize itself, and navigate autonomously. Furthermore, we present in-depth ablation studies of the important components of the system and their impact on localization accuracy. Videos, code, and additional experiments can be found on the project website: https://sites.google.com/view/low-cost-quadruped-slam
Paper Structure (15 sections, 15 equations, 6 figures, 3 tables)

This paper contains 15 sections, 15 equations, 6 figures, 3 tables.

Table of Contents

  1. INTRODUCTION
  2. Related works
  3. SYSTEM DESIGN
  4. Contact estimation
  5. Legged Odometry as Least Squares
  6. Scan Stabilization
  7. Integrating leg odometry with visual inerital odometry
  8. Integrating leg odometry with 2D SLAM
  9. Experimental Evaluation
  10. Experimental Setting
  11. Integrated Navigation Stack
  12. Mapping Results
  13. Robot autonomy
  14. Real robot experiments
  15. CONCLUSIONS

Figures (6)

  • Figure 1: Overall system design for robust navigation in low-cost quadruped robots. Our contributions are highlighted in light blue, different colors indicate different types of data.
  • Figure 2: Effects of scan stabilization in the map generation. Left: no scan stabilization. Right: with scan stabilization.
  • Figure 3: Estimated trajectories of the Silver badger robot in the small warehouse environment
  • Figure 4: Autonomous exploration mapping results.
  • Figure 5: Estimated trajectories using our system and the baseline of the Silver badger robot in a real-world environment (Environment 1, IAS Lab)
  • ...and 1 more figures