基于组合定位海量数据处理的铁路机车安全距离预警  

Early warning on safety distance of railway locomotive based on massive data processing of integrated positioning

作　　者：周小平[1,2] 刘祥磊[3] 

机构地区：[1]北京建筑大学北京市建筑安全监测工程技术研究中心 [2]中国人民大学信息学院 [3]北京建筑大学现代城市测量国家测绘局重点实验室

出　　处：《中国安全生产科学技术》2015年第10期103-109,共7页Journal of Safety Science and Technology

基　　金：地理空间信息工程国家测绘地理信息局重点实验室项目(201317);北京市高等学校青年英才资助项目(YETP1660)

摘　　要：针对我国铁路机车定位数据丢失时有发生的现象,阐述了以现有较成熟的GNSS/INS/DR(Global Navigation Satellite System/Inertial Navigation System/Dead Reckoning)组合定位装置构建铁路机车定位的系统架构。在此基础上,针对海量位置数据,引入Hadoop技术,给出了采用HBase存储海量位置信息数据的方法,采用MapReduce校正位置信息数据及安全距离计算的方法。经实验验证,该方案在需要大量数据存取的情况下,采用MapReduce进行分布式数据分析效率更高,扩展性更好。Aiming at the phenomenon that the loss of positioning data for railway locomotive in China have occurred frequently,it was illustrated to construct the system architecture of railway locomotive positioning by using the integrated positioning equipment of GNSS / INS / DR（ Global Navigation Satellite System / Inertial Navigation System /Dead Reckoning）,which is a comparatively mature technology now available. On this basis,aiming at the massive position data and introducing into the Hadoop technology,the methods to store the massive data of position information by applying HBase and correct the position information data and calculate the safety distance by applying MapReduce were put forward. The experimental verification showed that for this scheme,under the condition of needing massive data access,the distributed data analysis by applying MapReduce has higher efficiency and better expansibility.

关 键 词：海量数据 铁路机车 安全距离 组合定位 HADOOP 

分 类 号：X928[环境科学与工程—安全科学]