Vehicle State Recognition And Driving Behavior Assessment Based On VANET

Rapid development of the automotive industry and urbanization brings increasingly problems of traffic congestion and road safety. Developing vehicular Ad-Hoc networking technology(VANET) to promote the construction of intelligent transportation systems, has great significance in improving the traffic efficiency and transport safety. As the driver is still the main controller of vehicle safety, the research of vehicle networking technology should focus on helping the driver to maintain safer driving state.This paper studied the algorithms of the vehicle state recognition and methods of comprehensive assessment of driving behavior based on VANET. Vehicle-mounted unit realizes the recognition of vehicle motion state effectively, acquires other vehicle’s state information through vehicular ad-hoc network. We design algorithms of comprehensive assessment of the driver’s driving behavior based on historical traffic data. The quantitative assessment of the driver’s overall state and driving performance provides scientific basis for the driver to enhance his driving skill. The main works are summarized as follows:Firstly, we studied the technology of vehicle motion state identification and recognize the vehicle state through acceleration sensors and gyroscopes. We designed algorithms of the vehicle state identification and hazard classification and validated the algorithms through experimental data. We researched information broadcast mechanism within the ad-hoc network based on danger level state of vehicle status. The vehicle recognition system realizes vehicle status’ s real-time perception, valid identification and warning terminal within VANET.Secondly, comprehensive assessment of the driver’s driving behavior includes two dimensions of safe driving and economical driving. We select the speed, acceleration, angular velocity and degree roll as the indicators to establish safe driving evaluation model based on fuzzy membership degree. Ideal throttle opening on different speeds has been calculated through gray prediction algorithm. Model of economic driving has been bulit based on throttle opening. Comprehensive evaluation model of driver’s behavior based on membership of safety driving and economic driving has been established. The rationality and effectiveness of integrated assessment models has been verified by using mathematical model to solve and analysis the historical traffic data. At the end, we achieved software design of comprehensive assessment system of driving behavior.