Aircraft intelligent air combat decision-making has become a new research hotspot for military powers in the world. In order to solve the decision-making problem of unmanned aerial vehicles in one-to-one intelligent air combat, an unmanned fighter aircraft decision-making model is established, and the maneuvering decision-making instructions of the intelligent body are solved by deep learning algorithm. First, establish a UAV flight kinematics/dynamics model to form a one-on-one air combat confrontation situation. Secondly, establish a reward function that comprehensively considers the advantage of attack angle, speed, height and distance to guide the agent to converge to the optimal solution. Finally, based on the TD3 algorithm, the fully connected network constructs the agent value network and the target network to realize the agent's decision-making. The simulation results show that the agent can effectively perceive the air battlefield situation, maintain the battlefield advantage, and complete the strike against the target.
Intelligent Air Combat Maneuvering Decision Based on TD3 Algorithm
Lect. Notes Electrical Eng.
International Conference on Autonomous Unmanned Systems ; 2022 ; Xi'an, China September 23, 2022 - September 25, 2022
Proceedings of 2022 International Conference on Autonomous Unmanned Systems (ICAUS 2022) ; Kapitel : 98 ; 1082-1094
10.03.2023
13 pages
Aufsatz/Kapitel (Buch)
Elektronische Ressource
Englisch
Maneuvering-decision analysis for air-to-air combat
British Library Online Contents | 1997
|Research on Autonomous Maneuvering Decision of UAV in Close Air Combat
Springer Verlag | 2024
|Maneuvering decision in air combat based on multi-objective optimization and reinforcement learning
British Library Online Contents | 2018
|Numerical simulation of maneuvering combat aircraft
Tema Archiv | 2006
|