Multi-UAV Cooperative Multi-objective Task Allocation Based on Deep Reinforcement Learning

Jingyi Guo, Shunmin Li, Guanqun Wu, Aijun Li, Yong Guo

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

Abstract

This paper proposes a task allocation algorithm based on the combination of reinforcement learning and deep neural networks to address the problem of multi-UAV cooperative multi-objective task allocation. It utilizes graph neural networks (GNN) and attention mechanisms to model the policy function, thereby constructing a task allocation strategy based on the collective state of the UAV swarm which enables the reinforcement learning algorithm to generalize to varying numbers of enemy target nodes in the environment. To improve the efficiency and stability of training, the S-sample batch reinforcement learning algorithm is adopted. The simulation results demonstrate that the algorithm can effectively solve the multi-UAV task allocation problem.

Original languageEnglish
Title of host publicationAdvances in Guidance, Navigation and Control - Proceedings of 2024 International Conference on Guidance, Navigation and Control Volume 12
EditorsLiang Yan, Haibin Duan, Yimin Deng
PublisherSpringer Science and Business Media Deutschland GmbH
Pages341-352
Number of pages12
ISBN (Print)9789819622436
DOIs
StatePublished - 2025
EventInternational Conference on Guidance, Navigation and Control, ICGNC 2024 - Changsha, China
Duration: 9 Aug 202411 Aug 2024

Publication series

NameLecture Notes in Electrical Engineering
Volume1348 LNEE
ISSN (Print)1876-1100
ISSN (Electronic)1876-1119

Conference

ConferenceInternational Conference on Guidance, Navigation and Control, ICGNC 2024
Country/TerritoryChina
CityChangsha
Period9/08/2411/08/24

Keywords

  • Attention Mechanisms
  • Graph Neural Networks
  • Multi-UAV
  • Reinforcement Learning
  • Task Allocation

Fingerprint

Dive into the research topics of 'Multi-UAV Cooperative Multi-objective Task Allocation Based on Deep Reinforcement Learning'. Together they form a unique fingerprint.

Cite this