| In recent years,traffic accidents caused by Smombies crossing the roads have become a problem worthy of attention in society.The purpose of this study is to study the problem of Smombies crossing the road and propose solutions.Firstly,the simulation is used to quantify the impact of Smombies on traffic safety and efficiency,and then the Smombie situation is identified by multi-source information fusion context awareness.And we explored how to design the early warning mechanism based on Vehicle-to-Personal(V2P)communication mode in the future,and evaluated its effectiveness to provide auxiliary support for traffic managers in future policymaking.The research innovations of this paper are as follows:(1)A simulation model is constructed to quantify the influence of Smombie on mixed traffic flowAt present,some studies mainly focus on the impact on Smombie individuals,but not on group traffic.To quantitatively study the safety and efficiency effects of Smombies on group traffic,a hybrid traffic flow model with Smombie is constructed by using the cellular automata simulation method.The simulation rules of Smombie,a new traffic participant,are well defined based on their behaviors when crossing the roads.The Nasch and Blue simulation models are improved according to the humanvehicle interaction rules in mixed traffic flow.It is found that in terms of efficiency,the average speed of vehicles decreases with the increase of the proportion of Smombie in different traffic density and passenger density scenarios.Among them,the increase of Smombie has the most obvious effect on the average speed of“medium or high density traffic flow-medium density passenger flow ”,and the decrease ratio can reach more than 50%.In terms of safety,under different traffic density and passenger density scenarios,with the increase of the proportion of Smombie,the number of vehicle sharp brakes increases gradually,indicating that the more Smombie,the lower the safety.In the low traffic density scene,when the proportion of Smombie is 10%,the number of sharp brakes increases by 64% to147% compared with that of the No-Smombie scene.When the proportion of Smombie is 50%,the increase of the number of sharp brakes reaches 210%-411%,that is,twice to four times higher than that of No-Smombie scene.(2)An improved D-S evidence theory based method for Somobie context awareness is proposedAt present,the application of context awareness in the traffic field mainly focuses on driver’s driving behavior,but less on pedestrian traffic situation recognition.To identify the crossing road Smombie situation to intervene in early warning,this study proposes a crossing road Smombie context awareness framework based on multi-source information,and designs an improved sensor fusion algorithm based on D-S evidence theory and fuzzy mathematical theory.The algorithm can effectively identify the posture of mobile phones using when users cross the road.The accuracy and recall rate are above 0.9.(3)An early warning method for the Smombies crossing the road based on human-Vehicle communication mode is proposedAt present,the vehicles ’ avoidance of Smombie is mainly based on visual recognition,limited by the range of horizon,and the reaction time is very short,which is an important reason for the accident.Therefore,after identifying the Smombie crossing the road,this study designs an early warning mechanism based on the future human-vehicle communication mode,which sends early warning to pedestrians and vehicles respectively.It is proved by simulation that the early warning method can reduce the influence of Smombies on mixed traffic flow and improve safety greatly without losing much efficiency. |
PDF Full Text Request