Deep reinforcement learning for real-world quadrupedal locomotion: a comprehensive review
Problem
Building agile, intelligent controllers for four-legged (quadrupedal) robots, a long-standing challenge in robotics and AI.
Why it matters
Agile legged locomotion is described as one of the typical challenges in the pursuit of AI.
What the AI did
This is a review surveying deep reinforcement learning methods that learn locomotion controllers end-to-end without physical modeling, covering algorithms, simulation environments, hardware platforms and reality-gap solutions.
Results
not stated in source
Impact
Authors state DRL has achieved major breakthroughs in quadrupedal locomotion, though practical problems are alleviated only 'to a certain extent'.
Record
| Venue | Intelligence & Robotics |
|---|---|
| Published | 2022 |
| Work type | article |
| Authors | Hongyin Zhang, He Li, Donglin Wang |
| Citations | 19 (OpenAlex, at import time) |
| References | 119 |
| Model or method | Deep reinforcement learning (DRL); no specific algorithm named |
| Organization | not stated in source |
| Primary source | https://doi.org/10.20517/ir.2022.20 |
| Open access | hybrid |
| Keywords | quadrupedal locomotion, deep reinforcement learning, legged robots, reality gap, survey |
Importance rating
real but narrower advance
Abstract calls agile legged robot controllers one of the typical challenges in the pursuit of AI, with DRL achieving major breakthroughs in quadrupedal locomotion; a core robotics/AI research problem, but 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
|