GPS Data Analysis System For Taxi Based On Hadoop Technology

With the development of information technology,in the intelligent transportation industry,sensors,collection devices,universal assembly and other electronic collection technology development,traffic information data,the rapid surge in the amount of data.In the face of increasing traffic problems such as traffic congestion,unreasonable use of traffic resources and frequent traffic accidents,the traditional data analysis model can not use the large data information generated in real life to analyze and solve the contradiction of traffic problems in reality.Taxi GPS information is a typical traffic data information in large traffic data.Through analyzing the GPS data of a city taxi,it can get the traffic performance of the city directly and effectively.It can provide data support and strategy basis for resolving traffic contradiction,optimizing traffic resource utilization,formulating effective traffic management policy and providing reasonable travel suggestion by using the analysis conclusion of traffic characteristic.In this paper,the background and significance of the research are described,and the status quo of the research and analysis on GPS large data at home and abroad are introduced.The key technologies of HDFS,YARN,Sqoop and Spark,which are used in the construction and application of GPS large data analysis system of taxi,are expounded and the technical characteristics are analyzed.When building the system,it is necessary to build an open source Hadoop data processing platform.The Hadoop large data processing platform software and the components needed for the subsequent data analysis are chosen to demonstrate the feasibility of the system.Through cluster deployment and management,the taxi GPS Cluster management environment for large data analysis.The large data information is stored and managed by HDFS,the core component of Hadoop platform.Spark parallel computing framework preprocesses the GPS data and matches the GIS map and cleans the data to get the analyzed data source.From the passenger travel time,Passenger travel space from two angles,the use of clustering algorithm for taxi GPS data analysis of traffic characteristics,and ultimately use the Sqoop tool to analyze the results of the data derived and the results of the demonstration.Based on the analysis of taxi GPS data and the urban cultural information,the characteristics of passenger travel time and other characteristics of the distribution,as well as the taxi Empty-Loaded rate,a single trip consumption,the average travel time consumption,hot spots and hot spots Regional analysis and other characteristics of the city's traffic travel,traffic management departments can provide the basis for travel trips for the residents to make guidance.