Aarhus University Seal / Aarhus Universitets segl

Erdal Kayacan

Planning swift maneuvers of quadcopter using motion primitives explored by reinforcement learning

Publikation: Bidrag til bog/antologi/rapport/proceedingKonferencebidrag i proceedingsForskningpeer review

In this work, we propose a novel, learning-based approach for swift maneuver planning of unmanned aerial vehicles using motion primitives. Our approach is composed of two main stages: learning a set of motion primitives during offline training first, and utilization of them for online planning of fast maneuvers thereafter. We propose a compact disposition of motion primitives which consists of roll, pitch, and yaw motions to build up a simple yet effective representation for learning. Thanks to this compact representation, our method retains an easily transferable, reproducible, and referable knowledge which caters for real-time swift maneuver planning. We compare our approach with the current state-of-the-art methods for planning and control, and show improved navigation time performance up to 25 % in challenging obstacle courses. We validate our approach through software-in-the-loop Gazebo simulations and real flight tests with Diatone FPV250 Quadcopter equipped with PX4 FMU.

OriginalsprogEngelsk
TitelProceedings of the American Control Conference : 2019 American Control Conference, ACC 2019
Antal sider7
ForlagIEEE
Udgivelsesår2019
Sider279-285
Artikelnummer8815352
ISBN (Elektronisk)9781538679265
StatusUdgivet - 2019
Begivenhed2019 American Control Conference, ACC 2019 - Philadelphia, USA
Varighed: 10 jul. 201912 jul. 2019

Konference

Konference2019 American Control Conference, ACC 2019
LandUSA
ByPhiladelphia
Periode10/07/201912/07/2019
SponsorBoeing, et al., GE Research, General Motors co, Mitsubishi Electric Research Laboratory (MERL), United Technologies Research Center (UTRC)

Se relationer på Aarhus Universitet Citationsformater

ID: 172716335