航空学报 > 2026, Vol. 47 Issue (17): 333180-333180   doi: 10.7527/S1000-6893.2026.33180

基于舰面环境模型的舰载机保障作业调度算法

罗祎喆1,2,3, 王佳宝1, 余新得1, 陈旭东1, 金钊1,2,3(), 冯朔1,2,3, 石育澄1,2,3, 徐明亮1,2,3   

  1. 1.郑州大学 计算机与人工智能学院,郑州 450001
    2.智能集群系统教育部工程研究中心,郑州 450001
    3.国家超级计算郑州中心,郑州 450001
  • 收稿日期:2025-12-03 修回日期:2026-03-26 接受日期:2026-04-28 出版日期:2026-05-20 发布日期:2026-05-14
  • 通讯作者: 金钊 E-mail:jinzhao@zzu.edu.cn
  • 基金资助:
    国家自然科学基金(62406292);国家自然科学基金(62302459);国家自然科学基金(62406293);国家自然科学基金(62325602);国家自然科学基金(62036010)

Scheduling algorithm for carrier-based aircraft support operations based on deck environment model

Yizhe LUO1,2,3, Jiabao WANG1, Xinde YU1, Xudong CHEN1, Zhao JIN1,2,3(), Shuo FENG1,2,3, Yucheng SHI1,2,3, Mingliang XU1,2,3   

  1. 1.School of Computer and Artificial Intelligence,Zhengzhou University,Zhengzhou 450001,China
    2.Engineering Research Center of Intelligent Swarm Systems,Ministry of Education,Zhengzhou 450001,China
    3.National Supercomputing Center in Zhengzhou,Zhengzhou 450001,China
  • Received:2025-12-03 Revised:2026-03-26 Accepted:2026-04-28 Online:2026-05-20 Published:2026-05-14
  • Contact: Zhao JIN E-mail:jinzhao@zzu.edu.cn
  • Supported by:
    National Natural Science Foundation of China(62406292)

摘要:

在航母舰载机保障作业调度中,无模型强化学习(MFRL)在动态甲板场景下受到物理环境建模精度的制约,而基于模型的强化学习(MBRL)因环境模型与决策模型在迭代训练中存在相互依赖的协同优化问题,面临计算复杂度高与收敛困难的挑战。对此,提出了一种融合有模型与无模型(MB-MF)特性的混合强化学习框架。首先,利用历史调度数据训练基于深度神经网络的甲板环境模型,使其在最小容差范围内精确预测状态转移;然后,将收敛后的环境模型替代真实环境,内嵌入交互环境中,结合深度Q网络(DQN)算法训练调度智能体,实现环境模型学习与策略优化的解耦;最后,经实验验证表明,与使用物理环境的MFRL相比,该方法在无需精确建模的情况下性能差距仅为4%。而相较于MBRL基线方法,舰载机出动时间则缩短34%。同时在资源受限场景中,决策速度较启发式方法提高近300倍,而调度质量仅降低17%。

关键词: 舰载机, 保障作业, 调度优化, 强化学习, 深度学习

Abstract:

In aircraft carrier flight deck scheduling operations, Model-Free Reinforcement Learning (MFRL) is constrained by the precision of physical environment modeling under dynamic deck conditions. In contrast, Model-Based Reinforcement Learning (MBRL) faces challenges of high computational complexity and convergence difficulties due to the interdependent co-optimization between the environment model and the decision-making model during iterative training. To address these issues, This paper proposes a hybrid reinforcement learning framework that integrates Model-Based and Model-Free (MB-MF) characteristics. First, a deep neural network-based deck environment model is trained using historical scheduling data to accurately predict state transitions within a minimal tolerance range. Then, the converged environment model is embedded into the interactive environment in place of the real environment, where a scheduling agent is trained using the Deep Q-Network (DQN) algorithm, thereby decoupling environment model learning from policy optimization. Experimental results demonstrate that, compared to MFRL using the physical environment, the proposed method achieves a performance gap of only 4% without requiring precise modeling. Moreover, it reduces the aircraft sortie time by 34% relative to the MBRL baseline. In resource-constrained scenarios, the decision-making speed is nearly 300 times faster than that of heuristic methods, while the scheduling quality is reduced by only 17%.

Key words: carrier-based aircraft, deck operations, complex job scheduling, reinforcement learning, deep learning

中图分类号: