Obstacle Avoidance Drone by Deep Reinforcement Learning and Its Racing with Human Pilot
Problem
Training drones to avoid obstacles without the laborious labeled datasets that supervised learning requires.
Why it matters
Obstacle avoidance capability is central to safe and useful drone operation, including racing against human pilots.
What the AI did
Deep reinforcement learning agents in discrete and continuous action spaces, including a U-net segmentation model with an actor-critic network, were trained to avoid obstacles in three environments and race human pilots.
Results
Best continuous algorithm 'easily outperforms' (abstract truncated before full comparison); no numbers stated in source.
Impact
Demonstrated RL-trained drones performing obstacle avoidance and racing against human pilots.
Record
| Venue | Applied Sciences |
|---|---|
| Published | 2019 |
| Work type | article |
| Authors | Sangyun Shin, Yong-Won Kang, Yong-Guk Kim |
| Citations | 55 (OpenAlex, at import time) |
| References | 41 |
| Model or method | U-net segmentation with actor-critic network (deep reinforcement learning) |
| Organization | not stated in source |
| Primary source | https://doi.org/10.3390/app9245571 |
| Open access | gold |
| Keywords | drone, reinforcement learning, obstacle avoidance, actor-critic, U-net |
Importance rating
real but narrower advance
Obstacle avoidance is central to safe, useful drone operation; abstract cites research attention and racing human pilots but states no scale of impact.
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
|