Learning Perception-Aware Agile Flight in Cluttered Environments

Kód výsledku v IS VaVaI

<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F68407700%3A21230%2F23%3A00366389" target="_blank" >RIV/68407700:21230/23:00366389 - isvavai.cz</a>

Výsledek na webu

<a href="https://doi.org/10.1109/ICRA48891.2023.10160563" target="_blank" >https://doi.org/10.1109/ICRA48891.2023.10160563</a>

DOI - Digital Object Identifier

<a href="http://dx.doi.org/10.1109/ICRA48891.2023.10160563" target="_blank" >10.1109/ICRA48891.2023.10160563</a>

Jazyk výsledku

angličtina

Název v původním jazyce

Learning Perception-Aware Agile Flight in Cluttered Environments

Popis výsledku v původním jazyce

Recently, neural control policies have outperformed existing model-based planning-and-control methods for autonomously navigating quadrotors through cluttered environments in minimum time. However, they are not perception aware, a crucial requirement in vision-based navigation due to the camera's limited field of view and the underactuated nature of a quadrotor. We propose a learning-based system that achieves perception-aware, agile flight in cluttered environments. Our method combines imitation learning with reinforcement learning (RL) by leveraging a privileged learning-by-cheating framework. Using RL, we first train a perception-aware teacher policy with full-state information to fly in minimum time through cluttered environments. Then, we use imitation learning to distill its knowledge into a vision-based student policy that only perceives the environment via a camera. Our approach tightly couples perception and control, showing a significant advantage in computation speed ($10times$ faster) and success rate. We demonstrate the closed-loop control performance using hardware-in-the-loop simulation.

Název v anglickém jazyce

Learning Perception-Aware Agile Flight in Cluttered Environments

Popis výsledku anglicky

Recently, neural control policies have outperformed existing model-based planning-and-control methods for autonomously navigating quadrotors through cluttered environments in minimum time. However, they are not perception aware, a crucial requirement in vision-based navigation due to the camera's limited field of view and the underactuated nature of a quadrotor. We propose a learning-based system that achieves perception-aware, agile flight in cluttered environments. Our method combines imitation learning with reinforcement learning (RL) by leveraging a privileged learning-by-cheating framework. Using RL, we first train a perception-aware teacher policy with full-state information to fly in minimum time through cluttered environments. Then, we use imitation learning to distill its knowledge into a vision-based student policy that only perceives the environment via a camera. Our approach tightly couples perception and control, showing a significant advantage in computation speed ($10times$ faster) and success rate. We demonstrate the closed-loop control performance using hardware-in-the-loop simulation.

Druh

D - Stať ve sborníku

CEP obor

—

OECD FORD obor

20204 - Robotics and automatic control

Projekt

<a href="/cs/project/GM23-06162M" target="_blank" >GM23-06162M: TOPFLIGHT: Plánování trajektorií a misí agilních vzdušných robotů v prostředí s překážkami</a><br>

Návaznosti

P - Projekt vyzkumu a vyvoje financovany z verejnych zdroju (s odkazem do CEP)

Rok uplatnění

2023

Kód důvěrnosti údajů

S - Úplné a pravdivé údaje o projektu nepodléhají ochraně podle zvláštních právních předpisů

Název statě ve sborníku

2023 IEEE International Conference on Robotics and Automation

ISBN

979-8-3503-2365-8

ISSN

1050-4729

e-ISSN

2577-087X

Počet stran výsledku

7

Strana od-do

1989-1995

Název nakladatele

IEEE

Místo vydání

Piscataway

Místo konání akce

Londýn

Datum konání akce

29. 5. 2023

Typ akce podle státní příslušnosti

WRD - Celosvětová akce

Kód UT WoS článku

001036713001114