Vehicle Abnormal Driving Trace Model

With the rapid development of the economy and the constant growth of the vehicle population, the incidence of the traffic accident is high and the traffic accident deaths are on the raise. All of these bring a increasingly serious problem of the traffic safety. Now, the research on the traffic safety has become a research hotspot in the field of transportation. According to the statistical data, we find that most of the traffic accidents are caused by the abnormal vehicle behavior. Therefore, finding the method to quickly identify the abnormal vehicle driving behavior and warning the driver immediately has become an urgent problem to be resolved.The existing technologies to identify the abnormal vehicle driving behaviors detect it and issued a warning only by the road monitoring equipments or simply using the car navigation device. But there is no a technology identifying the behavior by vehicle itself and giving a warning immediately and the resulting message lag will cause the traffic accident. This paper researches and analyses the vehicle driving trace in the position of the vehicles and then creates the models of the vehicle abnormal driving trace. At last we do an experimental verification based on the smart car.Firstly, we research the vehicle abnormal driving behaviors and then classify them into two categories. Secondly, we do a lot researches and analysis on the vehicle driving trace according to the principles of Traffic Engineering and the numerical approximation theory, and then build the models of the abnormal vehicle driving trace. Thirdly we make a smart car using the STM32F103processor which is based on the ARM Cortex-M3core. We also used the μC/OS-II,the μC/GUI and the nRF(Radio Frequency) technology to make and control the smart car. At last, we build a simple experimental road network and verify the models for vehicle abnormal driving trace by test if the smart car can identify tow common abnormal driving behaviors of the retrogradation and the abnormal U-turn. By analyzing the results of the experiment, we conclude that the smart car can identify the abnormal driving behaviors correctly and give a real-time warning.This paper builds the models for abnormal vehicle driving trace by combining the theoretical analysis and experimental verification and proposes a method to identify the abnormal vehicle driving behaviors quickly and give a warning simultaneously. Then we verify the method through the models of vehicle abnormal driving trace. The results of the experiment show that the method is practicable. It is helpful to reduce the number of traffic accidents and improve traffic safety condition.