城市交通拥堵日益严重,高效的路径导航方法一直是当前研究的热点和缓解拥堵的主要途径.现有的研究成果主要集中在对单个车辆行驶时间的路径寻优和小规模路网的多车辆均衡化的路径导航,没有实现大规模多车辆多路径的实时动态路径导航.当...城市交通拥堵日益严重,高效的路径导航方法一直是当前研究的热点和缓解拥堵的主要途径.现有的研究成果主要集中在对单个车辆行驶时间的路径寻优和小规模路网的多车辆均衡化的路径导航,没有实现大规模多车辆多路径的实时动态路径导航.当前研究主要存在以下局限:(1)导航方案评价指标单一,不能充分表示导航方案的优劣;(2)无法实现大规模路网的实时导航.针对这些问题,本文提出一种城市交通路网实时动态多路口路径导航量子搜索方法(A Route Guidance Method based on Quantum Searching for Real-time Dynamic Multi-intersections in Urban Traffic Networks,RGQS),该方法充分考虑各种因素,实时提供大规模路网的路径导航.本文的实验分别在人工路网和真实路网中验证了RGQS方法相比于对比算法可以使行驶时间减少达到20%.展开更多
With the emerging connected autonomous driving paradigm,more advanced applications leveraging vehicular communications are drawing tremendous attentions.In order to analyze the feasibility and performance of these app...With the emerging connected autonomous driving paradigm,more advanced applications leveraging vehicular communications are drawing tremendous attentions.In order to analyze the feasibility and performance of these applications,it is necessary to build an evaluation platform that jointly considers vehicular communication,road traffic and vehicle dynamics.This article describes our recent progress on network-level autonomous driving simulator based on the Cellular-Vehicle-to-Everything(C-V2X)protocol,and a joint platform combined with SUMO and CARLA simulators for evaluating road traffic and vehicle dynamics.To demonstrate its effectiveness,this article implements a hybrid multi-intersection scheduling scheme on the platform,and shows the advantages of the scheme in terms of traffic efficiency and fault tolerance.A remote driving application based on CARLA,wherein the interplay between communication and computation is also investigated.展开更多
文摘城市交通拥堵日益严重,高效的路径导航方法一直是当前研究的热点和缓解拥堵的主要途径.现有的研究成果主要集中在对单个车辆行驶时间的路径寻优和小规模路网的多车辆均衡化的路径导航,没有实现大规模多车辆多路径的实时动态路径导航.当前研究主要存在以下局限:(1)导航方案评价指标单一,不能充分表示导航方案的优劣;(2)无法实现大规模路网的实时导航.针对这些问题,本文提出一种城市交通路网实时动态多路口路径导航量子搜索方法(A Route Guidance Method based on Quantum Searching for Real-time Dynamic Multi-intersections in Urban Traffic Networks,RGQS),该方法充分考虑各种因素,实时提供大规模路网的路径导航.本文的实验分别在人工路网和真实路网中验证了RGQS方法相比于对比算法可以使行驶时间减少达到20%.
基金This work was supported by the National Key R&D Program of China(Grant No.2019YFE0196600)the Nature Science Foundation of China(No.61871254,No.91638204,No.61861136003)the program for Professor of Special Appointment(Eastern Scholar)at Shanghai Institutions of Higher Learning,and research funds from the Shanghai Institute for Advanced Communication and Data Science(SICS).
文摘With the emerging connected autonomous driving paradigm,more advanced applications leveraging vehicular communications are drawing tremendous attentions.In order to analyze the feasibility and performance of these applications,it is necessary to build an evaluation platform that jointly considers vehicular communication,road traffic and vehicle dynamics.This article describes our recent progress on network-level autonomous driving simulator based on the Cellular-Vehicle-to-Everything(C-V2X)protocol,and a joint platform combined with SUMO and CARLA simulators for evaluating road traffic and vehicle dynamics.To demonstrate its effectiveness,this article implements a hybrid multi-intersection scheduling scheme on the platform,and shows the advantages of the scheme in terms of traffic efficiency and fault tolerance.A remote driving application based on CARLA,wherein the interplay between communication and computation is also investigated.
