殷书慧

个人信息

参与实验室科研项目

复杂环境下非完全信息博弈决策的智能基础模型研究

研究课题

针对不确定复杂环境下多群体博弈决策中的瓶颈问题，围绕其非完全信息、高智能、强动态的特点，从智能模型构建、多群体博弈决策理论形成以及人机对抗性能验证与评估等层面开展研究。

学术成果

共撰写/参与撰写专利 0 项，投出/录用/发表论文 1 篇。

Conference Articles

1. 1v1 Air Combat Maneuver Decision Using Minimax-DQN Shuhui Yin, Yu Kang, Yun-Bo Zhao , and Jian Xue In The 41st Chinese Control Conference 2022 [Abs] [PDF]

   Autonomous maneuver decision of UAV is the key to realize future air combat. Considering the strong antagonism and the uncertainty of opponents in air combat, traditional methods are difficult to solve Nash equilibrium strategy. This paper proposes a Minimax-DQN algorithm combining deep reinforcement learning with game theory, which solves the problem that the single agent reinforcement learning algorithm is difficult to converge due to the unstable environment. The effectiveness of the proposed algorithm is verified by the convergence test.