Dedicated left-turn lanes are traditionally used at intersections. This practice may not be optimal where heavy traffic exists from multiple directions. As is well known, the capacity can be increased if vehicles are grouped in the same direction in advance, but the additional infrastructure is usually needed, such as presignal systems. Fortunately, under the development of connected and automated vehicle (CAV) technology, the presignal system can be achieved by coordinating vehicles directly. To this end, we establish a modeling framework and management strategies for pure CAV traffic to achieve direction-based grouping. A rolling time-horizon–based sorting strategy is developed to achieve real-time coordination under uncertain vehicle arrivals. A series of numerical experiments have shown that the improvement of the proposed sorting strategy is significant and robust compared with traditional dedicated left-turn lane intersections without coordination, especially when the traffic is busy.

中文翻译：

---

基于滚动水平的信号交叉口完全协作交通策略

专用左转车道传统上用于十字路口。当来自多个方向的交通繁忙时，这种做法可能不是最佳做法。众所周知，如果提前将车辆按同一方向分组，可以增加容量，但通常需要额外的基础设施，例如预信号系统。幸运的是，在联网和自动驾驶汽车（CAV）技术的发展下，可以通过直接协调车辆来实现预信号系统。为此，我们建立了纯 CAV 流量的建模框架和管理策略，以实现基于方向的分组。开发了一种基于滚动时间范围的排序策略，以在不确定的车辆到达情况下实现实时协调。