Deep Reinforcement Learning for End-to-End Local Motion Planning of Autonomous Aerial Robots in Unknown Outdoor Environments: Real-Time Flight Experiments
Problem
Micro aerial vehicles must autonomously navigate obstacles and reach goals in unknown, GPS-denied outdoor environments despite limited size and computational power.
Why it matters
Collision-free autonomous flight in dynamic, GPS-denied settings is a core capability for aerial robots.
What the AI did
An actor–critic reinforcement learning policy was trained in a Gazebo simulator to map noisy MAV state and laser scans directly to continuous motion control, without mapping or path planning.
Results
Trained policy performs collision-free flight in the real world; no quantitative metrics stated in abstract.
Impact
Demonstrates simulator-trained RL transferring to real-world MAV point-goal navigation.
Record
| Venue | Sensors |
|---|---|
| Published | 2021 |
| Work type | article |
| Authors | Oualid Doukhi, Deok Jin Lee |
| Citations | 34 (OpenAlex, at import time) |
| References | 40 |
| Model or method | actor–critic deep reinforcement learning |
| Organization | not stated in source |
| Primary source | https://doi.org/10.3390/s21072534 |
| Open access | gold |
| Keywords | deep reinforcement learning, micro aerial vehicle, GPS-denied navigation, sim-to-real, collision avoidance |
Importance rating
real but narrower advance
Autonomous GPS-denied collision-free flight is a core aerial robotics capability demonstrated in real world; abstract states no scale of affected users or applications.
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
|