摘要
随着人工智能迅速发展以及"智慧机场"的提出,研究人工智能在机场如何有效地辅助机场管制人员,驾驶员指挥航空器在地面滑行具有重要意义。本文提出一种基于强化学习的滑行路径规划方法,构建航空器机场地面强化学习移动模型,并以海口美兰机场为案例采用Python内置工具包Tkinter进行场面仿真;在此基础上,考虑机场航空器滑行规则,采用Off-Policy中Q-Learning算法求解贝尔曼方程,实现航空器在Modelbased环境中进行静态路径规划。结果表明:本文所提方法能够实现停机位到跑道出口智能静态路径规划。
With the rapid development of artificial intelligence(AI)and the proposal of“smart airport”,it is of great importance to actively explore the application of AI in airports to assist airport controllers and pilots to com⁃mand aircraft to taxiing on the aircraft ground effectively.A taxiing path planning method based on reinforcement learning is proposed,a reinforcement learning mobile model of aircraft airport is constructed,and then Meilan Air⁃port of Haikou is taken as an example to achieve the scene simulation by using the Python built-in toolkit Tkinter.Considering the aircraft taxiing rules of the airport,the Q-Learning algorithm in Off-policy is used to solve the Bell⁃man equation to realize the AI static path planning of aircraft in the model-based environment.The results show that the proposed method can realize the AI static path planning of aircraft from gate position to runway exit.
作者
疏利生
李桂芳
嵇胜
SHU Lisheng;LI Guifang;JI Sheng(College of Civil Aviation,Nanjing University of Aeronautics and Astronautics,Nanjing 210016,China)
出处
《航空工程进展》
CSCD
2021年第3期65-70,共6页
Advances in Aeronautical Science and Engineering
