TY - GEN
T1 - Asynchronous Multi-agent Pareto Optimization for Diverse UAV Maneuver Strategy Generation
AU - Zhou, Tianze
AU - Zhang, Fubiao
AU - Sun, Zhiwen
AU - Liu, Mingcheng
AU - Wang, Zhaoshun
N1 - Publisher Copyright:
© 2023, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.
PY - 2023
Y1 - 2023
N2 - 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.
AB - 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.
KW - Maneuver strategy generation
KW - Multi-agent reinforcement learning
KW - Pareto optimization
UR - http://www.scopus.com/inward/record.url?scp=85151143767&partnerID=8YFLogxK
U2 - 10.1007/978-981-19-6613-2_120
DO - 10.1007/978-981-19-6613-2_120
M3 - Conference contribution
AN - SCOPUS:85151143767
SN - 9789811966125
T3 - Lecture Notes in Electrical Engineering
SP - 1209
EP - 1218
BT - Advances in Guidance, Navigation and Control - Proceedings of 2022 International Conference on Guidance, Navigation and Control
A2 - Yan, Liang
A2 - Duan, Haibin
A2 - Deng, Yimin
A2 - Yan, Liang
PB - Springer Science and Business Media Deutschland GmbH
T2 - International Conference on Guidance, Navigation and Control, ICGNC 2022
Y2 - 5 August 2022 through 7 August 2022
ER -