首页|接驳地铁的共享自动驾驶汽车动态合乘匹配研究

接驳地铁的共享自动驾驶汽车动态合乘匹配研究

扫码查看
为了实现共享自动驾驶汽车(Shared Autonomous Vehicles,SAV)和地铁的整合,提出采用SAV作为地铁"第一公里"(First Mile,FM)和"最后一公里"(Last Mile,LM)的接驳方式.构建了动态合乘匹配系统以厘清接驳地铁的SAV合乘的框架及运行规则,系统输入为路网数据、SAV数据(位置、载客状态等)、出行者数据(起点、终点等),输出为出行匹配结果、合乘路径.建立了出行路径模型和出行匹配模型来实现系统的合乘匹配功能.出行路径模型考虑出行者绕行时间最小化进行建模,可得到任意2名出行者合乘时的最优出行路径,用于出行匹配模型的参数计算.出行匹配模型协同考虑出行者服务质量和运营商利益以及SAV与地铁时刻表的联合优化进行建模,可得到最优出行匹配方案,包括人与SAV匹配、人与人匹配.选取纽约曼哈顿地铁进行4种出行需求(2 160、4 320、6 480、8 640次·h-1)的算例分析.结果表明:在出行需求2 160、4 320次·h-1的FM和LM出行、出行需求6 480次·h-1的LM出行中,出行者平均等待时间最大值为1.06 min,出行者平均绕行时间最大值为1.11 min,出行者平均匹配时长最长为46.09 s,每小时最多可为运营商节约877.24 km的车辆行驶里程,匹配率最高可达61.67%,系统平均求解时长最长为2.09 s;表明所建系统和模型具有高效性和稳定性.
Dynamic Ridesharing Matching of Shared Autonomous Vehicles as Connection to Metro
To integrate shared autonomous vehicles(SAV)with the metro,this study proposes the use of SAV as the connection of the metro's first mile(FM)and last mile(LM).A dynamic ridesharing matching system was constructed to clarify the framework and operating rules of SAV connected to the metro.The system inputs are the road network data,SAV data(location,passenger status,etc.),and traveler data(origin,destination,etc.),and the outputs are the matching results and travel paths.Travel path and travel matching models were developed to realize the ridesharing matching function of the system.The travel path model considers the minimization of the detour time of travelers,which can output the travel path of any two travelers when ridesharing,and was used for the parameter calculation of the travel matching model.The travel matching model considers the traveler service quality and operator benefits as well as the joint optimization of SAV and metro schedules to obtain the optimal travel matching scheme,including the matching of travelers to SAV and travelers to travelers.The metro in Manhattan,New York was selected for a case study at four travel demand scales(2 160,4 320,6 480,and 8 640 trips·h-1).The results show that for FM and LM travels with travel demands of 2 160 and 4 320 trips·h-1,as well as for LM travel with a travel demand of 6 480 trips·h-1,the maximum mean waiting time for travelers is 1.06 min;the maximum mean detour time for travelers is 1.11 min;the maximum mean matching time for travelers is 46.09 s;the maximum vehicle distance saved for the operator is 877.24 km·h-1;the maximum matching rate is 61.67%;and the maximum mean solution time for the system is 2.09 s.These results demonstrate the efficiency and stability of the system and model.

traffic engineeringshared autonomous vehiclesmetro connectiondynamic rideshar-ing matchingpath selection

霍月英、张越、李晓娟、陈国庆

展开 >

内蒙古大学交通学院,内蒙古呼和浩特 010070

内蒙古大学数学科学学院,内蒙古呼和浩特 010021

北京联合大学城市轨道交通与物流学院,北京 100101

交通工程 共享自动驾驶汽车 地铁接驳 动态合乘匹配 路径选择

国家自然科学基金项目国家自然科学基金项目内蒙古自治区关键技术攻关计划项目内蒙古自治区高等学校青年科技英才支持计划项目

52062039719610242019GG287NJYT23061

2024

中国公路学报
中国公路学会

中国公路学报

CSTPCD北大核心
影响因子:1.607
ISSN:1001-7372
年,卷(期):2024.37(6)
  • 4