Study On Urban Car Driving Characteristics Based On GPS Data

With the deeply research and development of Intelligent Transportation System (ITS), microscopic traffic simulation as one of its core content, is becoming a hot spots of traffic engineering research both at home and at abroad. The authenticity of behavior model deicide whether the simulation system is success or failure. Therefore, to study the characteristics of driver behavior and car operation in traffic flow is significant in microscopic traffic simulation.On the basis of project------"Data collection and analysis system based on GPS" ofBeijing Transportation R&D center, the paper has done the following work.(1) After a conclusion on the history of urban road traffic flow and driving behavior, this paper analyzes the development of driving behavior. The disadvantage of time series data of car following in existing research was found in this dissertation. The technology route was: starting of practice, depending on high precision instrument to collect car following data, exercising scientific theory methodology, combining with computer.(2) The paper has discussed the related Global Positioning System (GPS) technologies and the Geographical Information System (GIS) re-development technologies. Then some relative knowledge and key technologies of cloud theory, which has advantage on uncertainty knowledge representation Data Mining, was introduced in detail(3) Using GPS positioning data which was obtained in Beijing road network, the paper constructs a system based on the GPS/GIS to demonstrate the travel path of vehicles. The paper selected the urban expressway-Xitucheng road and urban arterial road-Xueyuan road in Beijing as research object. In the data processing procedure, the concept "short trip" was introduced. Firstly data was divided into data intervals and saved according to the concept "short trip". And then driving time and driving distance of the short trip was calculated. On this base the acceleration frequency of the vehicle on driving time and driving distance was calculated using the GPS data.(4) The paper analyses the traffic data in use of programming processing. From the microscopic angle the paper studies the vehicles driving behavioral trait on different road, compares the acceleration frequency of traveling time and distance in a short trip, study the velocity-time sequences graph and acceleration-time sequences graph of vehicle on urban expressway and urban arterial road main, analyses the characteristics of vehicle driving behaviors under different traffic conditions.(5) Using the advanced cloudy theory, the paper proposes an acceleration cloud model to establish the vehicles stable driving state, the accelerated state and the deceleration state.