Dream to Fly: Model-Based Reinforcement Learning for Vision-Based Drone Flight
Fuente:
arXiv
Gespeichert in:
| Hauptverfasser: | , , , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2025
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866908949924544512 |
|---|---|
| author | Romero, Angel Shenai, Ashwin Geles, Ismail Aljalbout, Elie Scaramuzza, Davide |
| author_facet | Romero, Angel Shenai, Ashwin Geles, Ismail Aljalbout, Elie Scaramuzza, Davide |
| contents | Autonomous drone racing has risen as a challenging robotic benchmark for testing the limits of learning, perception, planning, and control. Expert human pilots are able to fly a drone through a race track by mapping pixels from a single camera directly to control commands. Recent works in autonomous drone racing attempting direct pixel-to-commands control policies have relied on either intermediate representations that simplify the observation space or performed extensive bootstrapping using Imitation Learning (IL). This paper leverages DreamerV3 to train visuomotor policies capable of agile flight through a racetrack using only pixels as observations. In contrast to model-free methods like PPO or SAC, which are sample-inefficient and struggle in this setting, our approach acquires drone racing skills from pixels. Notably, a perception-aware behaviour of actively steering the camera toward texture-rich gate regions emerges without the need of handcrafted reward terms for the viewing direction. Our experiments show in both, simulation and real-world flight using a hardware-in-the-loop setup with rendered image observations, how the proposed approach can be deployed on real quadrotors at speeds of up to 9 m/s. These results advance the state of pixel-based autonomous flight and demonstrate that MBRL offers a promising path for real-world robotics research. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2501_14377 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Dream to Fly: Model-Based Reinforcement Learning for Vision-Based Drone Flight Romero, Angel Shenai, Ashwin Geles, Ismail Aljalbout, Elie Scaramuzza, Davide Robotics I.2.9; I.2.10; I.2.6 Autonomous drone racing has risen as a challenging robotic benchmark for testing the limits of learning, perception, planning, and control. Expert human pilots are able to fly a drone through a race track by mapping pixels from a single camera directly to control commands. Recent works in autonomous drone racing attempting direct pixel-to-commands control policies have relied on either intermediate representations that simplify the observation space or performed extensive bootstrapping using Imitation Learning (IL). This paper leverages DreamerV3 to train visuomotor policies capable of agile flight through a racetrack using only pixels as observations. In contrast to model-free methods like PPO or SAC, which are sample-inefficient and struggle in this setting, our approach acquires drone racing skills from pixels. Notably, a perception-aware behaviour of actively steering the camera toward texture-rich gate regions emerges without the need of handcrafted reward terms for the viewing direction. Our experiments show in both, simulation and real-world flight using a hardware-in-the-loop setup with rendered image observations, how the proposed approach can be deployed on real quadrotors at speeds of up to 9 m/s. These results advance the state of pixel-based autonomous flight and demonstrate that MBRL offers a promising path for real-world robotics research. |
| title | Dream to Fly: Model-Based Reinforcement Learning for Vision-Based Drone Flight |
| topic | Robotics I.2.9; I.2.10; I.2.6 |
| url | https://arxiv.org/abs/2501.14377 |