Learning quadrupedal locomotion on deformable terrain
Problem
Reinforcement learning control policies for legged robots fail on soft, deformable terrains, especially at high speed, because they cannot generalize beyond training data.
Why it matters
Robots that can walk reliably on sand, snow, or mud would expand rescue, inspection, and outdoor robotics.
What the AI did
Introduced a parameterized granular media model for reinforcement learning simulation and an adaptive control architecture that implicitly identifies terrain properties to boost locomotion performance on the in-house Raibo quadruped.
Results
not stated in source
Impact
Abstract claims improved locomotion on deformable terrain but is truncated before any quantitative results.
Record
| Venue | Science Robotics |
|---|---|
| Published | 2023 |
| Work type | article |
| Authors | Suyoung Choi, Gwanghyeon Ji, Jeongsoo Park, Hyeongjun Kim, Juhyeok Mun, Jeong Hyun Lee, Jemin Hwangbo |
| Citations | 175 (OpenAlex, at import time) |
| References | 39 |
| Model or method | not stated in source |
| Organization | not stated in source |
| Primary source | https://doi.org/10.1126/scirobotics.ade2256 |
| Open access | closed |
| Keywords | reinforcement learning, quadruped robot, deformable terrain, granular media, adaptive control |
Importance rating
real but narrower advance
Abstract identifies failure of RL control on soft, deformable terrain as a key limitation of legged robotics; significance for real-world robot deployment is inferred; no scale or beneficiaries stated.
More in Robotics & Autonomy
| # | Paper | Year | Citations | Importance |
|---|---|---|---|---|
| 1 | Autonomous robotic laparoscopic surgery for intestinal anastomosis | 2022 | 384 |
7
|
| 2 | Solving Rubik's Cube with a Robot Hand | 2019 | 569 |
6
|
| 3 | PaLM-E: An Embodied Multimodal Language Model | 2023 | 346 |
6
|
| 4 | Assistive Navigation Using Deep Reinforcement Learning Guiding Robot With UWB/Voice Beacons and Semantic Feedbacks for Blind and Visually Impaired People | 2021 | 48 |
6
|
| 5 | Towards Deep Radar Perception for Autonomous Driving: Datasets, Methods, and Challenges | 2022 | 154 |
6
|