基于飞行轨迹的飞机飞行异常检测算法
|
摘要
航空飞行中的容错率相对较低,无论是人为的疏忽还是仪器的短暂故障都有可能导致非常严重的后果,所以异常检测对于航空飞行来说至关重要;传统的航空异常检测缺乏关于飞行轨迹方面的探索,但是飞行轨迹是异常出现后最直观的体现方式,具有重要价值。传统的航空异常检测方法多数不能适用于大量数据,只能针对某项故障进行区段检测。由此入手,轨迹异常检测依据方向分布信息从粗到细针对轨迹点分析,避免大量的复杂异构数据处理,在保证检测正确率的同时提升效率。
Fault tolerant flight rate is relatively low,it is short for negligence or fault instruments are likely to lead to very serious consequences,so it is very important for flight anomaly detection; anomaly detection on the lack of traditional aviation flight path of exploration,but the flight trajectory is abnormal after the most intuitive way that is of great value.Most of the traditional aviation anomaly detection methods cannot be applied to a large number of data,only for fault detection.Thus,trajectory anomaly detection is based on the direction distribution information from coarse to fine for trajectory point analysis,avoiding a large number of complex heterogeneous data processing,while ensuring the detection accuracy and improving efficiency.
Fault tolerant flight rate is relatively low,it is short for negligence or fault instruments are likely to lead to very serious consequences,so it is very important for flight anomaly detection; anomaly detection on the lack of traditional aviation flight path of exploration,but the flight trajectory is abnormal after the most intuitive way that is of great value.Most of the traditional aviation anomaly detection methods cannot be applied to a large number of data,only for fault detection.Thus,trajectory anomaly detection is based on the direction distribution information from coarse to fine for trajectory point analysis,avoiding a large number of complex heterogeneous data processing,while ensuring the detection accuracy and improving efficiency.
引文
[1]J.G.Lee,J.Han,X.Li.Trajectory Outlier Detection:A Partition-and-Detect Framework.Washington:Computer Society,2008.
[2]X.Li,J.Han,S.Kim.ROAM:Rule-Based and Motif-Based Anomaly Dection in Massive Moving Object Data Sets.SIAM Press,2007.
[3]刘良旭,乔少杰,刘宾.V基于R-Tree的高效异常轨迹检测算法.软件学报,2009,24:26-35.
[4]朱明.数据挖掘[M].合肥:中国科学技术大学出版社,2002.
[2]X.Li,J.Han,S.Kim.ROAM:Rule-Based and Motif-Based Anomaly Dection in Massive Moving Object Data Sets.SIAM Press,2007.
[3]刘良旭,乔少杰,刘宾.V基于R-Tree的高效异常轨迹检测算法.软件学报,2009,24:26-35.
[4]朱明.数据挖掘[M].合肥:中国科学技术大学出版社,2002.