DOMA: Deep Smooth Trajectory Generation Learning for Real-Time UAV Motion Planning

Jin Yu, Haiyin Piao*, Yaqing Hou, Li Mo, Xin Yang, Deyun Zhou

*Corresponding author for this work

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

In this paper, we present a Deep Reinforcement Learning (DRL) based real-time smooth UAV motion planning method for solving catastrophic flight trajectory oscillation issues. By formalizing the original problem as a linear mixture of dual-objective optimization, a novel Deep smOoth Motion plAnning (DOMA) algorithm is proposed, which adopts an alternative layer-by-layer gradient descending optimization approach with the major gradient and the DOMA gradient applied separately. Afterwards, the mix weight coefficient between the two objectives is also optimized adaptively. Experimental result reveals that the proposed DOMA algorithm outperforms baseline DRL-based UAV motion planning algorithms in terms of both learning efficiency and flight motion smoothness. Furthermore, the UAV safety issue induced by trajectory oscillation is also addressed.

Original languageEnglish
Title of host publicationProceedings of the 32nd International Conference on Automated Planning and Scheduling, ICAPS 2022
EditorsAkshat Kumar, Sylvie Thiebaux, Pradeep Varakantham, William Yeoh
PublisherAssociation for the Advancement of Artificial Intelligence
Pages662-666
Number of pages5
ISBN (Electronic)9781577358749
DOIs
Publication statusPublished - 13 Jun 2022
Event32nd International Conference on Automated Planning and Scheduling, ICAPS 2022 - Virtual, Online, Singapore
Duration: 13 Jun 202224 Jun 2022

Publication series

NameProceedings International Conference on Automated Planning and Scheduling, ICAPS
Volume32
ISSN (Print)2334-0835
ISSN (Electronic)2334-0843

Conference

Conference32nd International Conference on Automated Planning and Scheduling, ICAPS 2022
Country/TerritorySingapore
CityVirtual, Online
Period13/06/2224/06/22

Fingerprint

Dive into the research topics of 'DOMA: Deep Smooth Trajectory Generation Learning for Real-Time UAV Motion Planning'. Together they form a unique fingerprint.

Cite this