Study On The Key Techonologies Of Traffic Accident Reconstruction And Precaution

Judging from the situation of traffic accidents at home and abroad, the overall trend ofroad traffic safety is favorable. However, there are still many problems waiting for people toconduct more researches, so as to effectively improve road traffic safety. Therefore,researches can be carried out in aspects of summarizing rules through analyzing thereconstruction of road traffic accidents and studying the techniques that can actively preventtraffic accidents. The present research mainly covers three aspects. The first is about trafficaccident reconstruction techniques. In this part, accident vehicle speed prediction model andmethod are proposed, and the scene trace information collection table is improved. Besides,Pc-Crash-based accident reconstruction simulation and modeling techniques are also studied.The second is about safety assessment techniques in the virtual road traffic environment, andboth subjective and objective assessment methods of road safety are suggested here. The thirdis about vehicle active safety techniques targeted for drivers. The main work andachievements of this research are as follows.(1) In terms of vehicle speed prediction in traffic accident reconstruction, athrow-distance-based accident vehicle speed prediction model, which was based on thepedestrian’s first falling site, was set up, and the accident vehicle speed prediction formulasfor vehicle-pedestrian collision accidents were deduced considering sites with and withoutslopes. The influence of road steps on accident vehicle speed prediction was also analyzed,and the matrix of relation between the accident vehicle speed and the throw distance (basedon the pedestrian’s first falling site) was provided. In addition, the graphing method wascombined with optimization algorithm, so as to present a vehicle speed prediction method forvehicle-pedestrian collision accidents. The model and methods not only laid down afoundation for reconstruction simulation, but also provided support for the police at site topredict the vehicle speed in the shortest time.(2) With regard to traffic accident scene trace analysis and information collection, traceanalyses were conducted from the perspectives of a single vehicle, multi-vehicles, as well asvehicle and pedestrians. Judgment methods for the scope of impact points in vehicle collision,collision angles, driving directions and pedestrian touch points were proposed, and the scenetrace was classified. The significance, timeliness and pertinence of making the scene traceinformation collection table were stated. Besides, a single vehicle accident was chosen as anexample to demonstrate the previous process in the aspects of accident morphologicalanalysis, accident reconstruction and table making, and the scene trace information collection table for single vehicle road traffic accidents was thus presented.(3) In terms of traffic accident simulation modeling and reconstruction, thePc-Crash-based modeling and reconstructing methods for vehicle-vehicle collision accidentsand vehicle sideslip accidents were studied. The vehicle speed calculation method based onthe monitoring video was applied to verify the simulation results of the Pc-Crash software,which was proved objective and credible. Upper and lower bound method, difference methodand response surface method were applied to check the uncertainty of the accidentreconstruction and simulation results. It was found that the adaptability of the responsesurface method was the best, with not many simulations and highly precise analysis results.(4) Considering road safety assessment, the road section traffic safety analysis methodbased on accident reconstruction was presented, and an accident in which a coach fell off thebridge was chosen to demonstrate the simulation flow and discuss accident prevention. Arapid generating system of the three-dimensional scene data was developed for constructingroad network. Moreover, subjective and objective assessment methods of road safety wereproposed in the virtual road traffic environment. They contributed greatly to the safetyassessment of uncompleted roads or road reconstruction schemes, and played a positive rolein preventing accidents from the source.(5) With regard to vehicle active safety techniques, drivers’ most typical behaviors ofcontrolling speed, keeping safe distance and controlling the steering wheel were considered,and with the driver-oriented design philosophy, the vehicle following model of theminimum safe distance was set up based on the environment where pedestrians, vehicles androads coexisted. The method of extracting the steering wheel boundary characteristics in a cabwith the integrated outline was proposed. Based on this, steering wheel positioning wasrealized, and both rough and fine AOI were established. Besides, the study of drivers’ illegaldriving behavior identification was carried out according to the hand characteristics of thedrivers. The achievements were laid down a good foundation for the subsequent intelligentanalysis of the abnormal behaviors of long-distance coach drivers.The methods and models mentioned above have been demonstrated and tested through alot of simulation experiments or actual cases. The results show that they are feasible in thefield of traffic safety. Relevant studies have laid down a good foundation for the theoreticaland technological development of traffic accident reconstruction and traffic safety guarantee.