The unmanned and intelligent ship-to-air defense system has emerged as a prominent development trend. Deep reinforcement learning is deemed applicable to combat command decision-making, offering potential to enhance combat effectiveness and reduce risk. However, there is a paucity of research on constructing intelligent models for ship-to-air defense problem in ship formation utilizing centroid jamming. To address this gap, we developed the two-dimensional model for centroid jamming scenario, and proposed a decision-making model based on the Markov decision-making process. This model aims to unify high-dimensional decision-making, encompassing the chaff cloud deployment and multi-ship maneuvering. Additionally, a threat level assessment model for enemy anti-ship missile is established to enhance the efficiency and success rate of the decision-making algorithm. Finally, the paper presents tests conducted on ship fleet of varying sizes and formations in diverse wind force environments, followed by an analysis of the results.
Decision-Making for Ship Formation Centroid Jamming Based on Reinforcement Learning
Lect. Notes Electrical Eng.
International Conference on Guidance, Navigation and Control ; 2024 ; Changsha, China August 09, 2024 - August 11, 2024
2025-03-11
11 pages
Article/Chapter (Book)
Electronic Resource
English
An Algorithm for Jamming Decision Using Dual Reinforcement Learning
British Library Online Contents | 2018
|Cognitive Electronic Jamming Decision-Making Method Based on Improved Q-Learning Algorithm
DOAJ | 2021
|Networked vehicle formation decision-making method based on nested graph reinforcement learning
European Patent Office | 2024
|