网联自动驾驶车辆通过信号交叉口的速度轨迹优化  

Speed Trajectory Optimization of Connected Autonomous Vehicles at Signalized Intersections

作　　者：陈壮壮 罗莉华[1,2] CHEN Zhuangzhuang;LUO Lihua(College of Transport and Communications,Shanghai Maritime University,Shanghai 201306,China;Transportation Research Institute,University of Michigan,Ann Arbor 48109,USA)

机构地区：[1]上海海事大学交通运输学院,上海201306 [2]美国密歇根大学交通科学研究所,美国安娜堡48109

出　　处：《交通信息与安全》2021年第4期92-98,156,共8页Journal of Transport Information and Safety

基　　金：上海市“科技创新行动计划”软科学重点项目(20692190700)资助。

摘　　要：以网联自动驾驶汽车(Connected Autonomous Vehicle,CAV)为研究对象,研究了CAV车队通过城市信号交叉口的速度轨迹优化控制策略。基于最优控制理论,采用CAV的自动驾驶模型描述车间相互作用,以所有CAV车辆在行驶过程中的总油耗为优化目标,根据信号灯的配时信息建立模型约束,通过优化CAV头车的速度轨迹,保证整个CAV车队在绿灯相位下快速通过交叉口并实现油耗最小。为了对该优化控制进行高效求解,采用离散Pontryagin极小值原理建立最优解的必要条件,利用基于神经网络训练的弹性反向传播(Resilient backpropagation,RPROP)算法设计了数值求解算法。多个典型场景的仿真结果显示:整个CAV车队均能在不停车的情形下通过信号交叉口,避免因在红灯时间窗到达停车线造成的停车、启动等过程,总油耗量最高可减少69.74%。该控制方法利用网联自动驾驶技术的优势,显著改善了城市交通通行效率和燃油经济性。The work studies the optimal control strategy of the speed trajectory for the connected autonomous vehicle(CAV)platoon at urban signalized intersections.Based on the optimal control theory,the automatic driving model is utilized to describe the interaction among vehicles.With the total fuel consumption for the CAV platoon considered as the optimization objective,the constraints of the model are established according to the timing phase of the traffic signal.All the CAVs in the platoon can pass through the intersection with the minimized total fuel consumption by optimizing the speed trajectory for the leading CAV.The necessary conditions for the optimal solution are obtained based on Pontryagin’s minimum principle to solve the proposed optimal controller.Then,the numerical solving algorithm is developed utilizing the resilient backpropagation(RPROP)solution algorithm.The simulation results for multiple representative scenarios show that the whole CAV platoon can pass through the signalized intersection without any stop and avoid stopping and starting caused by reaching the stop line at the red time window.Moreover,the total amount of fuel consumption can be decreased by 69.74%at most.The proposed method which takes advantage of CAV technology can improve the traffic efficiency and fuel economy for urban transportation.

关 键 词：交通控制 网联自动驾驶汽车 速度轨迹优化 最优控制 燃油经济性 

分 类 号：U491[交通运输工程—交通运输规划与管理]