Automated Vehicle Feasibility Study

This study collected automated vehicle (AV) performance data on public roadways in Athens, Ohio. The route for the study contained a combination of roads with different functional classifications, conditions, annual average daily traffic, and ownership responsibilities for maintenance and repair. Preparation for the public road deployment was done in a controlled environment at Transportation Research Center’s SMARTCenter, a dedicated AV test facility in East Liberty, Ohio. Researchers analyzed data and extracted insights relevant for both AV developers and infrastructure owners and operators.

The study found that rural environments offer a unique set of roadway features such as hills and curves, which can challenge the driving behavior of an AV. Rural regions can also contain a large number of low-traffic gravel roads that lack pavement markings, which appear to be a crucial infrastructure element for operation of current generation AVs. Similarly, the presence of well-maintained lane lines along curves can influence the AV’s roadway departure tendencies. The study found that curvature-related behavior of an AV is also influenced by driving speed on the roadway segment. Such findings were consistent regardless of the time of day along the route or season of data collection. However, commentary about AV performance in active adverse weather cannot be made, as this is still an area of active research.

本研究收集了美国俄亥俄州雅典市公共道路上的自动驾驶汽车（Automated Vehicle，AV）性能数据。本次研究的测试路线涵盖了功能分类、路况、年平均日交通量及养护维修责任主体各不相同的多种道路组合。本次公共道路实车部署的前期筹备工作，于俄亥俄州东自由市交通研究中心（Transportation Research Center）旗下的专属自动驾驶汽车测试设施SMARTCenter内的受控环境中完成。研究人员对采集到的数据开展分析，并提取出对自动驾驶汽车开发者、基础设施所有者及运营方均具有参考价值的核心结论。

研究发现，乡村道路环境具备一系列独特的道路特征，例如坡道与弯道，这些特征会对自动驾驶汽车的驾驶行为构成挑战。乡村区域还存在大量低流量的砂石路面，且未施划路面标线；而路面标线是当前世代自动驾驶汽车运行所需的关键基础设施要素。与之类似，弯道沿线养护完好的车道线，会对自动驾驶汽车的车道偏离倾向产生影响。研究还表明，自动驾驶汽车与曲率相关的驾驶行为，同样会受对应路段行驶速度的影响。上述研究结论不受路线行驶时段或数据采集季节的影响，均保持一致。不过，由于恶劣天气场景下的自动驾驶汽车性能仍属于活跃研究领域，本次研究无法针对该场景下的车辆表现作出评述。