基于车联网V2P 的行人碰撞风险辨识研究

为探索基于车联网V2P(Vehicle to Pedestrian)通信技术的行人碰撞风险辨识方法， 首先，在车联网环境下实时获取了目标位置、速度、运动方向等信息，并分析了典型人-车相 对运动场景中交通参与者的行为不确定性，进而提出了人-车碰撞区域随机几何模型；然后， 综合考虑了车联网系统的通信延时、定位误差、人-车相对运动不确定性等多因素的影响，建 立了人-车碰撞事故概率和冲突风险程度模型；最后，通过仿真实验分析了行车速度、通信延 时、定位精度等因素对行人碰撞风险辨识模型效果的影响，以及各因素间的相关性关系.本文 提出的方法对行人安全保护研究具有一定的参考价值，研究结果同时指出了车联网系统通信 延时与定位精度的技术要求.

Vehicle-pedestrian Collision Risk Assessment Based on Connected Vehicle V2P Communication

In this paper, a novel method is investigated for assessing vehicle- pedestrian collision risk in road traffic on the basis of connected vehicle V2P (Vehicle to Pedestrian) communication. First of all, a general V2P communication scenario is constructed to enable pedestrian motion being detected by approaching vehicles, explicitly along with real time obtaining the objects trajectory, velocity, orientation etc., while the typical behaviors of involved vehicles and pedestrians are analyzed, which in return, a stochastic geometric model is established for explicitly describing the pedestrian and vehicle location distribution in near crash situations. Then, vehiclepedestrian crash probability and risk evaluation model is established with comprehensively considering the V2P communication delay, positioning accuracy, uncertainty of vehicle- pedestrian relative motion. Finally, simulated Connected Vehicle test is conducted to examine the performance of pedestrian crash assessment model under the influence of connected vehicle communication delay, positioning accuracy and vehicle speed, and specially explore the relativity among these factors. The proposed method provides reference value for practical pedestrian safety application. The research results also indicate the technical requirements of Connected Vehicle system for future safety application.