文章编号： 1672-5328（2020）02-0118-09

龚龑，鲁啸
(北京市科学技术情报研究所，北京100044)

摘要： 在普通汽车与自动驾驶车辆混合行驶的道路网中，自动驾驶车辆视觉系统的辨识能力对交通 安全至关重要。根据自动驾驶车辆视觉系统基本结构，分析其与人视觉的能见度差距。模拟驾驶人 调速算法被证明在恶劣的天气下(包括低能见度时)优于自动驾驶车辆的两个典型调速算法，但是分 析发现自动驾驶车辆采用模拟驾驶人调速算法可能引发交通事故。因此，提出基于信息物理系统 (CPS)的安全调速算法，解决了自动驾驶车辆和驾驶人驾驶普通汽车之间视觉差距引发的交通安全 问题。为比较这两种算法的安全性，建立低能见度时视觉差距导致交通事故的道路网模型。该模型 分三个阶段描述了从导致交通事故至局部道路网瘫痪的全过程。最后，建立道路网事故发生概率和 平均瘫痪时间两个指标。仿真结果表明，自动驾驶车辆采用模拟驾驶人调速算法引发交通事故率较 高，而采用基于CPS的安全调速算法则不会发生交通事故。

关键词： 城市交通；自动驾驶；车辆视觉系统；信息物理系统(CPS)；调速算法；道路网模型；低 能见度

中图分类号： U491.1+2

文献标识码：A

Speed Regulation on Automatic Vehicles Based on Cyber-Physical System

Gong Yan, Lu Xiao
(Beijing science and technology information research institute, Beijing 100044, China)

Abstract： In the roadway network where the common vehicles and the automatic vehicles travel together, the recognition capability of the automatic vehicles' vision system is important to travel safety. According to the basic structure of vision system located on automatic vehicles, this paper analyzes the visibility gap between it and human vision. The simulation driver's speed regulation algorithm has been proved to be superior to the two typical speed regulation algorithms in severe weather (including low visibility). However, the analysis shows that the simulation driver's speed regulation algorithm may cause traffic accidents. Therefore, based on Cyber- Physical System (CPS), the paper proposes a speed regulation algorithm to solve the traffic safety problem caused by the visual gap between the automatic vehicles and common vehicles. In order to compare the safety of the two algorithms, the paper develops a roadway network model of traffic accidents caused by visual gap in low visibility. The model can describe the entire process from the occurrence of a traffic accident to local roadway network paralysis in three stages. Finally, a probability index of roadway network accidents and an index of the average paralysis time are established. The simulation results show that the traffic accident rate of the simulated driver by the speed regulation algorithm is high, while the traffic accident would not happen by the safety speed regulation algorithm based on CPS.

Keywords： urban transportation; automatic driving; vehicle vision system; Cyber-Physical Systems (CPS); speed regulation algorithm; roadway network model; low visibility