Learning to walk in confined spaces using 3D representation
Authors:
- 13 May 2024
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 8649-8656
Abstract
Legged robots have the potential to traverse complex terrain and access confined spaces beyond the reach of traditional platforms thanks to their ability to carefully select footholds and flexibly adapt their body posture while walking. However, robust deployment in real-world applications is still an open challenge. In this paper, we present a method for legged locomotion control using reinforcement learning and 3D volumetric representations to enable robust and versatile locomotion in confined and unstructured environments. By employing a two-layer hierarchical policy structure, we exploit the capabilities of a highly robust low-level policy to follow 6D commands and a high-level policy to enable three-dimensional spatial awareness for navigating under overhanging obstacles. Our study includes the development of a procedural terrain generator to create diverse training environments. We present a series of experimental evaluations in both simulation and real-world settings, demonstrating the effectiveness of our approach in controlling a quadruped robot in confined, rough terrain. By achieving this, our work extends the applicability of legged robots to a broader range of scenarios.Keywords
All Related Versions
This publication has 33 references indexed in Scilit:
- Lifelike agility and play in quadrupedal robots using reinforcement learning and generative pre-trained modelsNature Machine Intelligence, 2024
- Neural Volumetric Memory for Visual Locomotion Control2023
- Elevation Mapping for Locomotion and Navigation using GPU2022
- Vision-Guided Quadrupedal Locomotion in the Wild with Multi-Modal Delay Randomization2022
- Autonomous Spot: Long-Range Autonomous Exploration of Extreme Environments with Legged Locomotion2020
- Perceptive whole‐body planning for multilegged robots in confined spacesJournal of Field Robotics, 2020
- Vision Aided Dynamic Exploration of Unstructured Terrain with a Small-Scale Quadruped Robot2020
- Voxblox: Incremental 3D Euclidean Signed Distance Fields for on-board MAV planning2017
- Adaptive Motion Planning for Autonomous Rough Terrain Traversal with a Walking RobotJournal of Field Robotics, 2015
- Onboard perception-based trotting and crawling with the Hydraulic Quadruped Robot (HyQ)2013