飞机相遇模型仿真工具的研究

A research on Encountering-aircraft Trajectory Generator Tool

飞机的自主控制方法是空管领域研究的重点,尤其是飞机自主控制中的碰撞避险系统。本文提出了基于贝叶斯网络的飞机相遇模型,并以此模型为基础生成随机的相遇仿真数据。在实现仿真的过程中,本文利用有限的中国民航飞机飞行数据,先仿真一架飞机的航迹,在此基础上仿真两架相遇飞机每秒的高度、速度、加速度、转角速率大小,利用飞机相遇模型来依概率产生模拟相遇数据。此仿真航迹是没有任何碰撞预警情况下的航迹,两架飞机在相距一定距离之外均不会自主做出回避。本文利用了Matlab作为开发平台。

Autopilot is always the focus of aircraft control system, especially the automatical collision avoidance system. However, since aircraft collision seldom happens, it is usually insufficient to only use real-time encounter data to assess various collision avoidance system. We propose an aircraft encounter model based on Bayesian network. Then this encounter model is used to generate aircraft encounter data. We first simulate one aircraft trajectory, followed by aircraft encounter simulation, which includes the altitude, velocity, acceleration, and turn-rate by statistical probability. It is necessary to mention that our model considers no air traffic control intervention. Thus two pilots will not change their original flight intents before model. Our program use Matlab language. they enter the encounter cylinder, which is presumed by the encounter