自适应巡航及协同式巡航对交通流的影响分析

自适应巡航（ACC）和协同式自适应巡航（CACC）等自动驾驶技术正逐渐进入市场，未来一段时间内道路交通流将由人工驾驶车辆与不同等级、不同形式的自动驾驶车辆混合构成。为分析ACC和CACC对交通流的影响，利用实测交通数据NGSim建立人工驾驶车辆跟驰模型，并在综合已有ACC和CACC模型的基础上，提出基于安全间距的自动驾驶跟驰行为模型，进而得出不同ACC，CACC车辆渗透率下交通流的基本图模型。研究结果表明：自动驾驶可以提升交通容量；与ACC车辆比例r_a相比，CACC车辆比例r_c对交通容量的影响更为显著；当r_c>0.5时，饱和流量快速增加，当r_c=1时，饱和流量约为纯人工驾驶时的2倍。进一步，通过仿真考察车辆在车队中的跟驰响应和交通流在瓶颈处的运行情况。研究结果表明：自动驾驶改善了交通流的动态特性，对存在跟驰关系的连续车流来说，自动驾驶使得后车可以更加及时地响应前车的行为，车流会在更短的时间内进入稳态；在交通瓶颈处，自动驾驶降低了拥堵程度，提高了阻塞发生的临界流量。总体来看，自动驾驶对交通流静态和动态性能均有所提升，特别是在协同式自动驾驶场景下，车辆行为更加协调一致，交通流表现出良好的抗扰性，进一步验证了车路协同对自动驾驶的意义。

Effects of Adaptive Cruise Control and Cooperative Adaptive Cruise Control on Traffic Flow

Adaptive cruise control (ACC) systems and cooperative adaptive cruise control (CACC) systems are entering the market gradually. In the future, traffic flow will consist of manual vehicles and automated vehicles at different automation levels. To analyze the effect of ACC and CACC vehicles on traffic flow, this study established a car-following model for manual driving based on NGSim, which was obtained from real traffic flow observations; proposed a car-following model of ACC and CACC vehicles based on safety distance; and obtained fundamental diagrams for different penetration rates of ACC and CACC vehicles. The results showed that automated driving can improve traffic capacity. Compared to the penetration rate of ACC vehicles (r_a), the penetration rate of CACC vehicles (r_c) has a more significant impact on capacity. When r_c>0.5, the saturated capacity increases rapidly, and when r_c=1, the saturated capacity is approximately twice that of pure manual driving. Platoon responses and traffic flows at bottlenecks were also investigated. Results showed that automated driving enhances the dynamic characteristics of traffic flow. As for platoons, automated driving enables a following vehicle to respond to the preceding vehicle's behavior in time, and traffic flow reaches steady state in a shorter time. At bottlenecks, automated driving reduces congestion and increases the critical flow of congestion. In general, automated driving improves the static and dynamic performance of traffic flow, particularly in cooperative driving scenarios, where vehicle behavior is more consistent and traffic flow shows good stability, which further validates the significance of vehicle-infrastructure cooperation.