Asynchronous Multi-agent Pareto Optimization for Diverse UAV Maneuver Strategy Generation

Tianze Zhou, Fubiao Zhang*, Zhiwen Sun, Mingcheng Liu, Zhaoshun Wang

*Corresponding author for this work

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

Abstract

Recent advances have witnessed that Multi-Agent Reinforcement Learning (MARL) makes significant progress in Multi-UAV maneuver strategy generation. Difference from traditional MARL tasks, Multi-UAV combat scenarios are always in high dynamism and complexity, and exploring varying available maneuver strategies is necessary. In this paper, to extend the diverse maneuver strategy, we formalize the problem as the multi-objective optimization problem and propose an asynchronous Pareto-based multi-agent population optimization method. Besides, we propose the tolerance method to alleviate the Pareto front shock problem in the asynchronous Pareto optimization process. Finally, a 2V2 6-DOF UAV simulation environment is designed to evaluate the performance of the proposed methods. Experimental results show that our method can efficiently learn multiple maneuver strategies, such as counterattack and defense penetration.

Original languageEnglish
Title of host publicationAdvances in Guidance, Navigation and Control - Proceedings of 2022 International Conference on Guidance, Navigation and Control
EditorsLiang Yan, Haibin Duan, Yimin Deng, Liang Yan
PublisherSpringer Science and Business Media Deutschland GmbH
Pages1209-1218
Number of pages10
ISBN (Print)9789811966125
DOIs
Publication statusPublished - 2023
EventInternational Conference on Guidance, Navigation and Control, ICGNC 2022 - Harbin, China
Duration: 5 Aug 20227 Aug 2022

Publication series

NameLecture Notes in Electrical Engineering
Volume845 LNEE
ISSN (Print)1876-1100
ISSN (Electronic)1876-1119

Conference

ConferenceInternational Conference on Guidance, Navigation and Control, ICGNC 2022
Country/TerritoryChina
CityHarbin
Period5/08/227/08/22

Keywords

  • Maneuver strategy generation
  • Multi-agent reinforcement learning
  • Pareto optimization

Fingerprint

Dive into the research topics of 'Asynchronous Multi-agent Pareto Optimization for Diverse UAV Maneuver Strategy Generation'. Together they form a unique fingerprint.

Cite this