Multi-target interception scheme decision based on DDPG
Year: 2025
Type: conference-paper
Abstract: Deep reinforcement learning technology has the advantages of fast solving speed, strong model generalization ability, and good solving effect, so it is widely used in various disciplines and scenarios. In the air defense scenario, how to generate and optimize the interception scheme according to the currently observable information is very important. The generation of the existing interception schemes mostly depends on the mathematical model of artificial experience, and the solution effect and solution time are difficult to meet the actual needs when facing the increasingly complex and intelligent environment. To solve the above problems, combined with the specific requirements of specific scenarios, this paper adopts the Deep Deterministic Policy Gradient (DDPG) to optimize the interception scheme and verifies the effectiveness of the DDPG algorithm through simulation. The simulation results show that DDPG has faster decision-making speed and better solution effect. Finally, some suggestions for future algorithm improvement are given.
Authors: Hanwen Liu, Zhimin Zhuo, Xue Yang
Language: English
Topic: Guidance and Control Systems
Subfield: Aerospace Engineering
Field: Engineering
Domain: Physical Sciences
Open Access status: diamond
OpenAlex ID: W4408613117