您的位置:首页  > 论文页面

基于改进的DDPG机场机位分配算法研究

发表时间:2021-06-19  浏览量:159  下载量:45
全部作者: 顾存昕,周洪涛
作者单位: 华中科技大学人工智能与自动化学院
摘 要: 综合考虑机场的多个约束条件,以最大出港靠桥的航班数作为优化目标建立相应的数学模型,并将其转化成马尔可夫决策过程模型。设计环境的状态空间和智能体的动作空间,将大规模的离散动作空间通过构建特征的方式转变为连续动作空间,提出基于K最近邻(K nearest neighbor,KNN)和深度确定性策略梯度(deep deterministic policy gradient,DDPG)的机位分配算法,即DDPG_KNN. 以乌鲁木齐地窝堡国际机场的实际航班数据进行仿真实验来验证模型的有效性,所改进的算法能够提高机位资源的利用率。在对比实验中,DDPG_KNN的效果优于遗传算法。
关 键 词: 人工智能;机位分配;深度强化学习;DDPG
Title: Research on airport gate assignment based on improved DDPG algorithm
Author: GU Cunxin, ZHOU Hongtao
Organization: School of Artificial Intelligence and Automation, Huazhong University of Science and Technology
Abstract: Considering the multiple constraints of the airport comprehensively, a mathematical model is established by taking the maximum number of flights approaching the airport as the optimization objective, and the model is transformed into a Markov decision process model. The state space of the environment and the action space of the agent are designed, and the large-scale discrete action space is transformed into a continuous action space by the way of constructing features. An airport gate assignment algorithm based on K nearest neighbor (KNN) and deep deterministic policy gradient (DDPG) is proposed, namely DDPG_KNN. The actual flight data of Urumqi Diwopu international airport is used for simulation experiments to verify the effectiveness of the model, which can improve the utilization rate of gate resources. In the comparison experiment, the effect of DDPG_KNN is better than that of genetic algorithm.
Key words: artificial intelligence; airport gate assignment; deep reinforcement learning; DDPG
发表期数: 2021年6月第2期
引用格式: 顾存昕,周洪涛. 基于改进的DDPG机场机位分配算法研究[J]. 中国科技论文在线精品论文,2021,14(2):187-201.
 
1 评论数 0
暂无评论
友情链接