| Advanced traffic management system is the most basic part of the IntelligentTransportation System. It is a system which can monitor traffic flow real-timely,control intelligently and manage actively. It refers to the application of advancedcomputer technology, communication technology and sensor technology, and makethe vehicle, road and traffic management integration. Traffic information acquisitionis the critical foundation of the Advanced Traffic Management System. we can inferthe current traffic status of the road and predict the degree of congestion in the future,by achieving the traffic information which including traffic flow, vehicle speed, trafficdensity, share, queue length and so on with magnetic induction technology. In thispaper, the application of magnetic induction technology in Intelligent TransportationSystems is as the main object of study. It is mainly about two magnetic inductiondetection technology including induction coil and geomagnetic sensor, which areapplied in the traffic information collection and transportation system identification.Main studies and results of this dissertation are as following:Firstly, this paper designed an induction coil vehicle classifier which is based on theAVR microcontroller embedded systems. In the hardware circuit design, we useinexpensive microcontroller to replace the expensive industrial computer as the dataprocessing unit, which not only reduces costs, but also makes the detection devicecompact and easy to install. To solve the problem of the base frequency drift oninductive loop, a base frequency drift removal method based on sliding-time-windowclustering analysis has been presented. A method of vehicle recognition based on theK means clustering and Bayesian classifier is discussed both from the calculationspeed and learning training time. And experiment has verified that it is a simple, fastand efficient algorithm.Secondly, taking the conditions of road construction, environmental adaptability andanti-interference ability into account, this paper designs a wireless traffic detectionsystem based on geomagnetic sensor. The core design of the detection systemincludes geomagnetism detection module, wireless network communication module and traffic state identification module. For the stability problem of wirelesscommunication, it proposes cluster-type network topology with route forwarding,which using zigbee protocol, and adds custom handshake protocol to ensure theaccuracy and stability of the data communication.Thirdly, according to the characteristics of the curves which are detected by two kindsof magnetic induction sensor, a method of traffic state identification based onmagnetic induction curve sequence is discussed in this paper. It firstly extractscharacteristics from the curves which collected by the magnetic induction sensor, andthen determines the road traffic state by BP neural network classifier.This paper mainly introduces the application of magnetic induction detectiontechnology in the traffic system identification, analysis of the existing problems, andproposes creative solutions and algorithms. Experiments have verified that it iseffective and correct, and has broad application prospects. |
