基于自然轨迹考虑车型的多车道公路交通指标分析

Analysis of Traffic Indicators on Multi-Lane Highway Considering Car-Truck Interaction Based on Naturalistic Trajectory

分析了HighD和NGSIM两个开放的自然驾驶数据集中交通流参数(如速度和流量)与安全指标(如车头空距(DHW)、车头时距(THW)和碰撞时间(TTC))的异同。首先针对两个数据集分析了交通状态、不同车道和车辆类型等参数对交通流参数的影响。然后,针对车辆跟驰情况,研究了基于不同车道和不同车型的安全指标分布,探讨了安全指标与车辆行驶速度的关系。结果表明:(1)各车道平均车速与最小DHW和THW呈显著正相关,速度分布受交通状态、不同车道和车型的影响;(2)不同车道、不同车型对DHW和THW分布有显著影响;(3)THW适合于高速场景下跟驰状态的安全评价,THW和TTC的组合可以用来对低速场景进行安全评价。

The analysis and comparison of traffic indicators in different countries is essential for the design and development of traffic simulators and autonomous vehicle control strategy. In the paper, the similarities and differences between traffic flow parameters(such as speed and flow) and safety indicators such as headway(DHW), time headway(THW) and time to collision(TTC) in the two open natural driving data sets of HighD and NGSIM are analyzed. First, the traffic state of two data sets are introduced and the influence of various parameters, such as traffic state, different lanes and vehicle type on traffic flow parameters are analyzed. Then, for the car-following situation, safety indicators distribution based on different lanes and vehicle types are investigated, and the relationship between safety indicators and velocity is explored.The results show that:1)the average vehicle speed of each lane is significantly positively correlated with the minimum DHW and THW,and the speed distribution is affected by traffic conditions,different lanes and vehicle types;2)different lanes and different vehicle types have a significant impact on the distribution of DHWand THW;and 3)THWis suitable for the safety evaluation of the carfollowing state in high-speed scenario,and the combination of THW and TTC can be used for safety evaluation of low-speed scenario.