Research On Driver’s Preaccident Behavior Pattern Based On The Risk Time

Currently, the research on driver behaviour is a hot topic in the area of worldintelligent vehicle and intelligent transportation. Lack of actual accident data, thisresearch in China is still in the initial stage. This paper studies the driver behaviourbefore the crash accident of car and bicycle. It is intended to explore the relationshipbetween driver behaviour and accident risk. Setting the risk time as the researchobjective, this paper studies the factors that influence the risk time in thedriver-road-environment closed loop system. Moreover, the relationship of risk timeand bicycle driver injured risk is formulated and the braking model is built.Choosing30crash accidents data of car and bicycle from the database of In-depthInvestigation of Vehicle Accident in Changsha of China, this paper carries out theaccident reconstruction in PC-Crash and Madymo to extract the abstract of driverbehaviour. According to the difference of risk time in different car accidents, it isproposed that the risk time is influenced by road condition, time, weather, etc..However, the subjective factor of driver has marginal influence on it. Additionally, it isreported that the risk time and reaction time has simple linear relationship. Therefore,regarding risk time as the evaluation criteria has its own merit. It can abandon driver’ssubjective factor and well uncover a serious of driver’s reaction before car accident.There are some relationship between bicycle driver’s injured risk, vehicle speedand the risk time. To be specific, in low vehicle speed bicycle driver’s injured risk ismainly influenced by impact orientation and tumbling manner, and vehicle speed haslittle influence on it. However, in high vehicle speed, they are directly proportionrelated. This paper establishes the bicycle driver accident risk predicted model usingvehicle speed and risk time as the impact factors. Additionally, the method of buildingdriver braking model is discussed based on risk time. According to the analysis ofaccident data, the relationship of braking acceleration and risk time is researched.The result of this research can be served as a data support to the design ofvehicle active safety technology, driver behaviour intervention and the developmentof intelligent transportation. It can also be applied to the authentication and settlementof car accident, thereby improving car accident authentication level.