|
Decision support based on deep reinforcement learning in air combat maneuvers
|
Mahdi Zolfaghari   , Hojat Noormohammadi * , Gevork B. Gharehpetian |
| Skill National University |
|
|
Abstract: (14 Views) |
In recent years, group and collective air battles using drones have become popular in the armed forces of developed countries. In this kind of battles, drones are expected to perform agile and safety maneuvers according to the dynamic and complex environment of the battlefield. Deep reinforcement learning (DRL) technique, which is a suitable tool for continuous decision making in dynamic environments, will be a suitable option for decision support in air combat maneuvers (ACMD). Despite many studies in recent years on the functions of reinforcement learning in decision support, so far no road map has been drawn for using this technique in air combat decision support. Therefore, in this article, first, a complete picture of the issues in this field will be drawn. First, deep reinforcement learning will be described, and then its function in air combat decision support will be explained. Then a one-to-one decision making technique for winning drones in aerial combat maneuvers will be presented. Finally, the design and construction of the model and its evaluation have also been implemented.
|
|
| Keywords: Deep reinforcement learning, aerial maneuvering, decision support |
|
|
Full-Text [PDF 1370 kb]
(6 Downloads)
|
Type of Study: Research |
Subject:
Command
Received: 2023/05/12 | Accepted: 2023/08/19 | Published: 2026/08/22
|
|
|
|
|
|
|
| Send email to the article author |
|
|
|
| Add your comments about this article |
|
|
|
