Mastering air combat game with deep reinforcement learning

Zhu, Jingyu; Kuang, Minchi; Zhou, Wenqing; Shi, Heng; Zhu, Jihong; Han, Xu

doi:10.1016/j.dt.2023.08.019

Cited by 2 publications

(2 citation statements)

References 21 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Sun et al [12] applied self-play technique and observed emergence of multiple interpretable tactics. Piao et al [13] combined DRL with expert knowledge, and Zhu et al [14] designed an interruption mechanism where the agent increased decision frequency when facing emergency.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Interpretable DRL-based Maneuver Decision of UCAV Dogfight

Han,

Cheng,

2024

Preprint

View full text Add to dashboard Cite

This paper proposes a three-layer unmanned combat aerial vehicle (UCAV) dogfight frame where Deep reinforcement learning (DRL) is responsible for high-level maneuver decision. A four-channel low-level control law is firstly constructed, followed by a library containing eight basic flight maneuvers (BFMs). Double deep Q network (DDQN) is applied for BFM selection in UCAV dogfight, where the opponent strategy during the training process is constructed with DT. Our simulation result shows that, the agent can achieve a win rate of 85.75% against the DT strategy, and positive results when facing various unseen opponents. Based on the proposed frame, interpretability of the DRL-based dogfight is significantly improved. The agent performs yo-yo to adjust its turn rate and gain higher maneuverability. Emergence of "Dive and Chase" behavior also indicates the agent can generate a novel tactic that utilizes the drawback of its opponent.

show abstract

Section: Introductionmentioning

confidence: 99%

“…However, most works concerning DRL-based dogfight [9]- [14] only focused on simplified UCAV dynamics with three degrees of freedom (DOF), which is far from the real situation. Moreover, most works only claimed their algorithms outperformed conventional methods.…”

Section: Introductionmentioning

confidence: 99%