An air combat maneuver decision-making approach using coupled reward in deep reinforcement learning

Abstract In the domain of unmanned air combat, achieving efficient autonomous maneuvering decisions presents challenges. Deep Reinforcement learning(DRL) is one of the approaches to tackle this problem. The final performance of the DRL algorithm is directly affected by the design of the reward funct...

Full description

Saved in:
Bibliographic Details
Main Authors: Jian Yang, Liangpei Wang, Jiale Han, Changdi Chen, Yinlong Yuan, Zhu Liang Yu, Guoli Yang
Format: Article
Language:English
Published: Springer 2025-06-01
Series:Complex & Intelligent Systems
Subjects:
Online Access:https://doi.org/10.1007/s40747-025-01992-9
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849341656564760576
author Jian Yang
Liangpei Wang
Jiale Han
Changdi Chen
Yinlong Yuan
Zhu Liang Yu
Guoli Yang
author_facet Jian Yang
Liangpei Wang
Jiale Han
Changdi Chen
Yinlong Yuan
Zhu Liang Yu
Guoli Yang
author_sort Jian Yang
collection DOAJ
description Abstract In the domain of unmanned air combat, achieving efficient autonomous maneuvering decisions presents challenges. Deep Reinforcement learning(DRL) is one of the approaches to tackle this problem. The final performance of the DRL algorithm is directly affected by the design of the reward functions. However, the performance and convergence speed of the models suffer from unreasonable reward weights. Therefore, a method named Coupled Reward-Deep Reinforcement Learning(CR-DRL) is introduced to deal with this problem. Specifically, we propose a novel coupled-weight reward function for DRL within the air combat framework. The novel reward function integrates angle and distance so that our DRL maneuver decision model can be trained faster and perform better compared to that of the models use conventional reward functions. Additionally, we establish a brand new competitive training framework designed to enhance the performance of our model against personalized opponents. The experimental results show that our CR-DRL model outperforms the traditional model that uses the fixed-weight reward functions in this training framework, with a 6.3% increase in average reward in fixed scenarios and a 22.8% increase in changeable scenarios. Moreover, the performance of our model continually improves with the increase of iterations, ultimately yielding a certain degree of generalization performance against similar opponents. Finally, we develop a simulation environment that supports real-time air combat based on Unity3D, called Airfightsim, to demonstrate the performance of the proposed algorithm.
format Article
id doaj-art-84b47cfc62f94e5189799a8c9ab797cf
institution Kabale University
issn 2199-4536
2198-6053
language English
publishDate 2025-06-01
publisher Springer
record_format Article
series Complex & Intelligent Systems
spelling doaj-art-84b47cfc62f94e5189799a8c9ab797cf2025-08-20T03:43:34ZengSpringerComplex & Intelligent Systems2199-45362198-60532025-06-0111811710.1007/s40747-025-01992-9An air combat maneuver decision-making approach using coupled reward in deep reinforcement learningJian Yang0Liangpei Wang1Jiale Han2Changdi Chen3Yinlong Yuan4Zhu Liang Yu5Guoli Yang6College of Automation Science and Engineering, South China University of TechnologyCollege of Automation Science and Engineering, South China University of TechnologyCollege of Automation Science and Engineering, South China University of TechnologyCollege of Automation Science and Engineering, South China University of TechnologySchool of Electrical Engineering, Nantong UniversityCollege of Automation Science and Engineering, South China University of TechnologyDepartment of Big Data Intelligence, Advanced Institute of Big DataAbstract In the domain of unmanned air combat, achieving efficient autonomous maneuvering decisions presents challenges. Deep Reinforcement learning(DRL) is one of the approaches to tackle this problem. The final performance of the DRL algorithm is directly affected by the design of the reward functions. However, the performance and convergence speed of the models suffer from unreasonable reward weights. Therefore, a method named Coupled Reward-Deep Reinforcement Learning(CR-DRL) is introduced to deal with this problem. Specifically, we propose a novel coupled-weight reward function for DRL within the air combat framework. The novel reward function integrates angle and distance so that our DRL maneuver decision model can be trained faster and perform better compared to that of the models use conventional reward functions. Additionally, we establish a brand new competitive training framework designed to enhance the performance of our model against personalized opponents. The experimental results show that our CR-DRL model outperforms the traditional model that uses the fixed-weight reward functions in this training framework, with a 6.3% increase in average reward in fixed scenarios and a 22.8% increase in changeable scenarios. Moreover, the performance of our model continually improves with the increase of iterations, ultimately yielding a certain degree of generalization performance against similar opponents. Finally, we develop a simulation environment that supports real-time air combat based on Unity3D, called Airfightsim, to demonstrate the performance of the proposed algorithm.https://doi.org/10.1007/s40747-025-01992-9Air combatManeuver decision-makingDeep reinforcement learning(DRL)Coupled reward
spellingShingle Jian Yang
Liangpei Wang
Jiale Han
Changdi Chen
Yinlong Yuan
Zhu Liang Yu
Guoli Yang
An air combat maneuver decision-making approach using coupled reward in deep reinforcement learning
Complex & Intelligent Systems
Air combat
Maneuver decision-making
Deep reinforcement learning(DRL)
Coupled reward
title An air combat maneuver decision-making approach using coupled reward in deep reinforcement learning
title_full An air combat maneuver decision-making approach using coupled reward in deep reinforcement learning
title_fullStr An air combat maneuver decision-making approach using coupled reward in deep reinforcement learning
title_full_unstemmed An air combat maneuver decision-making approach using coupled reward in deep reinforcement learning
title_short An air combat maneuver decision-making approach using coupled reward in deep reinforcement learning
title_sort air combat maneuver decision making approach using coupled reward in deep reinforcement learning
topic Air combat
Maneuver decision-making
Deep reinforcement learning(DRL)
Coupled reward
url https://doi.org/10.1007/s40747-025-01992-9
work_keys_str_mv AT jianyang anaircombatmaneuverdecisionmakingapproachusingcoupledrewardindeepreinforcementlearning
AT liangpeiwang anaircombatmaneuverdecisionmakingapproachusingcoupledrewardindeepreinforcementlearning
AT jialehan anaircombatmaneuverdecisionmakingapproachusingcoupledrewardindeepreinforcementlearning
AT changdichen anaircombatmaneuverdecisionmakingapproachusingcoupledrewardindeepreinforcementlearning
AT yinlongyuan anaircombatmaneuverdecisionmakingapproachusingcoupledrewardindeepreinforcementlearning
AT zhuliangyu anaircombatmaneuverdecisionmakingapproachusingcoupledrewardindeepreinforcementlearning
AT guoliyang anaircombatmaneuverdecisionmakingapproachusingcoupledrewardindeepreinforcementlearning
AT jianyang aircombatmaneuverdecisionmakingapproachusingcoupledrewardindeepreinforcementlearning
AT liangpeiwang aircombatmaneuverdecisionmakingapproachusingcoupledrewardindeepreinforcementlearning
AT jialehan aircombatmaneuverdecisionmakingapproachusingcoupledrewardindeepreinforcementlearning
AT changdichen aircombatmaneuverdecisionmakingapproachusingcoupledrewardindeepreinforcementlearning
AT yinlongyuan aircombatmaneuverdecisionmakingapproachusingcoupledrewardindeepreinforcementlearning
AT zhuliangyu aircombatmaneuverdecisionmakingapproachusingcoupledrewardindeepreinforcementlearning
AT guoliyang aircombatmaneuverdecisionmakingapproachusingcoupledrewardindeepreinforcementlearning