Research On Traffic Safety Warning Method Based On Internet Of Vehicles

Traffic accident has become an important factor affecting people's normal life because of the huge casualties and economic losses.If the risk could be recognized and the driver could be warned timely,the traffic accidents are reduced effectively and traffic safety is improved.The Internet of Vehicles connected the vehicle and vehicle,vehicle and infrastructures with wireless technology,all the data can be shared in the system.The driver's cognitive ability is enhanced.These techniques offer new opportunities to real-time collision warning.A risk recognition and real-time collision warning method was proposed to assist the drivers,which aimed to improve the traffic safety.Firstly,Newell's low order car-following model was analyzed,the following vehicle's trajectory was obtained by shifted leading vehicle' trajectory according to driver's reaction time and minimum space.Based on the coordinates,vehicle's speed-space relationship could be established.Based on the relationship,the risk recognition method was proposed and the safety threshold of space could be obtained.If the real-time space less than the threshold,the vehicle may suffer risks.On the other hand,the probability of accident was calculated by piecewise function.The method was validated by analyzing the data offered by NGSIM project;the results indicated that the method could be more accurate description of the risk and possibility of the accident at the road scene.Secondly,for the complexity of accidents at the intersections,a temporal-spatial risk recognition method was proposed.On the basis of time region of vehicle gone through the conflict zone,the algorithm calculated probability of accident at unsignalized intersection was established.Due to the deficiencies of considering time coordinate only,spatial conflict which means the relationship among safety braking distance,reaction time and speed was used to correct the algorithm.A four-lane intersection was taken as an example,the comparison of results before and after the correction showed that this method could be more accurate description of the risk and the possibility of accidents at unsignalized intersection scene.Thirdly,the composition and interaction of real-time warning system were researched.According to analysis of existing warning mode,a warning mode combining visual and auditory was put forward,and the interface could display risk state continuously was designed,it could provide references for practical application.The warning issue was regarded as a decision-making problem,safety and signal disturbance were used as indexes,and the likely outcomes of different scheme were discussed and quantified by internal numbers method.A lot of work was done on the study of utility theory,and the utility function was build to represent the satisfaction about driving safety and noninterference.According to utility maximization and driver characteristics,different warning thresholds were determined.The results indicated that the method could improve the efficiency of warning.Finally,the simulation system was established for road and intersection scenes,a large number of experiments were done to verify the risk recognition and warning method proposed in this study.Based on the simulation parameters analysis,the simulation time interval 0.1s was determined.Experiments on warning efficiency indicated that the method could reduced the traffic accidents;the contrast experiments on missing warning and incorrect warning with time to collision(TTC)method showed that the method could reduce the rate of missing warning and incorrect warning;experiments on traffic severity expressed that the method could decrease the crash speed,so the accident severity was reduced.In conclusion,on the one hand,the method based on Newell's car-following model and temporal-spatial method in the intersection were effective for vehicles' risk recognition,and the possibility was expressed by probability;on the other hand,collision warning method with utility could improve the traffic safety.The thought was verified by those results.