异常轨迹数据预警与预测关键技术综述  

Key Technologies of Precaution and Prediction of Abnormal Spatial-Temporal Trajectory: A Review of Recent Advances

作　　者：仇功达 何明 杨杰[3] 曹玉婷 孙继红 

机构地区：[1]中国人民解放军陆军工程大学指挥控制工程学院,南京210002 [2]军事科学院系统工程研究院网络信息研究所,北京100071 [3]江苏省公安厅科技信息化处,南京210007 [4]南京市明基医院,南京210004

出　　处：《系统仿真学报》2017年第11期2608-2617,共10页Journal of System Simulation

基　　金：江苏省自然科学基金(BK20150721;BK20161469);江苏省科技基础设施建设计划(BM2014391);江苏省重点研发计划(BE2015728;BE2016904);国家重点研发计划(2016YFC0800606);中国博士后基金(2015M582786;2016T91017)

摘　　要：重大突发事件的事后处置已经愈加无法满足当前社会的迫切需求,急于需要向事前异常行为的预警预测转型。传感器网络与定位技术的快速发展与普及,为时空轨迹数据挖掘奠定了基础。围绕异常轨迹预警预测挖掘这一核心目的,对异常轨迹聚类识别与轨迹预测的国内外研究现状和进展进行了理论梳理、剖析,综述了相关算法在城市异常轨迹数据预警预测中的应用,指出了所面临的挑战和进一步的发展方向,为该领域的进一步研究提供参考。The ex-post disposition of a major incident, which is expected to transform into prediction and precaution of abnormal behavior, is increasingly unable to meet the urgent needs of the society.Therapid development and popularization of sensor network and positioning technology lay the foundation for mining spatial-temporal trajectory data. With the key objective of prediction and precaution of abnormal trajectory based on big data mining, the future research directions and prospects on trajectory clustering and recognitionareanalyzed, discussed and elaboratedinthis paper. Temporal trajectory prediction applied in prediction and precaution of abnormal spatial-temporal trajectory is also presented, providing a reference for further research on this field.

关 键 词：时空轨迹 异常行为 聚类 预测 

分 类 号：TP311[自动化与计算机技术—计算机软件与理论]