Kerner Boris S
Physics of Transport and Traffic, University of Duisburg-Essen, 47048 Duisburg, Germany.
Phys Rev E. 2023 Jul;108(1-1):014302. doi: 10.1103/PhysRevE.108.014302.
We have found that phase transitions occurring between three traffic phases [free flow (F), synchronized flow (S), and wide moving jam (J)] determine the spatiotemporal dynamics of traffic consisting of 100% automated-driving vehicles moving on a two-lane road with an on-ramp bottleneck. This means that three-phase traffic theory is a common framework for the description of traffic states independent of whether human-driving or automated-driving vehicles move in vehicular traffic. To prove this, we have studied automated-driving vehicular traffic with the use of classical Helly's model [Proceedings of the Symposium on Theory of Traffic Flow (Elsevier, Amsterdam, 1959), pp. 207-238] widely applied for automated vehicle motion. Although dynamic rules of the motion of automated-driving vehicles in a road lane are qualitatively different from those of human-driving vehicles, we have revealed that traffic breakdown (F→S transition) at the bottleneck exhibits the nucleation nature, which was observed in empirical field data measured in traffic consisting of 100% human-driving vehicles. The physics of the nucleation nature of the F→S transition in automated-driving traffic is associated with a discontinuity in the rate of lane-changing that causes the discontinuity in the rate of over-acceleration. This discontinuous character of over-acceleration leads to both the existence and self-maintaining of synchronized flow at the bottleneck in automated-driving vehicular traffic as well as to the existence at any time instant of a range of highway capacities between some minimum and maximum capacities. Within the capacity range, an F→S transition can be induced; however, when the maximum capacity is exceeded, then after some time-delay a spontaneous F→S transition occurs at the bottleneck. The phases F, S, and J can coexist each other in space and time.
我们发现，在三个交通相位[自由流（F）、同步流（S）和宽移动拥堵（J）]之间发生的相变决定了由在具有匝道瓶颈的双车道道路上行驶的100%自动驾驶车辆组成的交通的时空动态。这意味着三相交通理论是描述交通状态的通用框架，与车辆交通中是人类驾驶车辆还是自动驾驶车辆行驶无关。为了证明这一点，我们使用了广泛应用于自动驾驶车辆运动的经典赫利模型[《交通流理论研讨会论文集》（爱思唯尔，阿姆斯特丹，1959年），第207 - 238页]来研究自动驾驶车辆交通。尽管自动驾驶车辆在道路车道上的运动动态规则与人类驾驶车辆的规则在性质上有所不同，但我们已经揭示，瓶颈处的交通拥堵（F→S转变）具有成核性质，这在由100%人类驾驶车辆组成的交通中实测的经验现场数据中也观察到了。自动驾驶交通中F→S转变的成核性质的物理原理与换道速率的不连续性有关，这种不连续性导致了过度加速速率的不连续性。这种过度加速的不连续特性导致了自动驾驶车辆交通瓶颈处同步流的存在和自我维持，以及在任何时刻都存在一定范围的高速公路通行能力，介于某些最小和最大通行能力之间。在通行能力范围内，可以诱导F→S转变；然而，当超过最大通行能力时，经过一段时间延迟后，瓶颈处会自发发生F→S转变。相位F、S和J可以在空间和时间上相互共存。
