Home // SIMUL 2025, The Seventeenth International Conference on Advances in System Modeling and Simulation // View article

Decision Modeling for Unmanned Swarm Suppression of Enemy Air Defenses Based on Deep Reinforcement Learning

Authors:
Xiao Hu
Yonglin Lei
Fusong Luo
Hongfei Shi
Jiajun Zhu

Keywords: UAV swarm ;SEAD; decision-making modeling; combat simulation ; Dueling DQN

Abstract:
The Suppression of Enemy Air Defense (SEAD) mission is a critical component of Unmanned Aerial Vehicle (UAV) swarm operations, presenting a complex challenge for modeling and simulation. Machine Learning (ML), particularly Deep Reinforcement Learning (DRL), offers a promising approach to enhance UAV swarm SEAD effectiveness through intelligent decision-making. This paper, therefore, explores a modeling and simulation approach to intelligent combat equipment decision-making based on deep DRL. We establish a DRL modeling framework grounded in combat simulation and specifically construct an intelligent decision-making framework for UAV Swarm SEAD. Focusing on the attack decision-making problem, we present a case study utilizing the Dueling Deep Q-Network (Dueling DQN) algorithm for intelligent combat decision modeling. Preliminary experimental results demonstrate that the ML-based intelligent decision-making model achieves superior combat effectiveness compared to traditional knowledge engineering-based models.

Pages: 15 to 24

Copyright: Copyright (c) IARIA, 2025

Publication date: September 28, 2025

Published in: conference

ISSN: 2308-4537

ISBN: 978-1-68558-300-2

Location: Lisbon, Portugal

Dates: from September 28, 2025 to October 2, 2025