基于车路协同的智能车调度系统及方法研究

发布时间：2018-03-18 06:34

本文选题：车路协同　切入点：智能车　出处：《大连理工大学》2014年硕士论文　论文类型：学位论文

【摘要】：随着全球城市的大力发展,土地资源越来越紧张,汽车保有量持续增加,城市交通拥堵问题日益突出,智能交通运输系统的相关研究越来越受到各国的重视。作为智能交通运输系统的重要组成部分,车路协同系统将车辆与道路通过计算机互联网连接起来,旨在解决交通拥堵,提高道路利用率等问题。随着现代计算机及互联网科技的持续革新,智能无人驾驶车技术越来越成熟。本文研究的智能车辆调度系统,旨在为车路协同系统与智能车两者提供一种可行的合作方案,在未来车路协同大环境下,实现智能车的自动无冲突安全行驶,减少交通拥堵,提高道路交通利用率。本文研究了基于改进Dijkstra算法的单元目标最优路径规划方法,为每个运行目标提供一条初始的最优运行路线,使得车辆的行驶路径长度最短。本改进算法针对经典Dijkstra算法在任务节点增加时时间复杂度增加、实时性降低这一问题,提出了一种动态路径规划算法,它能动态地添加需求的任务节点,剔除冗余的任务节点,从而保持路网总节点数不变,进而降低了算法的时间复杂度。采用VC编写了算法仿真软件,对两种算法进行了比较,证明了改进算法的高效性。在单元目标路径规划的基础上,本文提出了基于全局离线规划和局部在线相结合的综合路径冲突解决方案,当多目标初始运行时采用全局规划,为每个目标提供一条初始的最优运行路线,在运行过程中采用Petri网建立冲突模型进行冲突车辆的局部规划,并开发了相关的测试软件验证了方案的可行性。最后,搭建和设计了硬件试验平台,采用STM32微处理器设计了路面基站模型,并采用相应的无线收发模块发送和接收来自模型智能车辆的数据信息并处理,证明了基于车路协同的智能车调度的可行性。为未来智能交通运输系统的发展提供了参考。
[Abstract]:With the rapid development of global cities, land resources are becoming more and more tight, the number of cars is increasing, and the problem of urban traffic congestion is becoming increasingly prominent. As an important part of intelligent transportation system, vehicle-road cooperative system connects vehicles and roads through the computer Internet to solve traffic congestion. With the continuous innovation of modern computer and Internet technology, the technology of intelligent driverless vehicle becomes more and more mature. The purpose of this paper is to provide a feasible cooperative scheme for the vehicle-road coordination system and intelligent vehicle. In the future environment of vehicle-road coordination, the intelligent vehicle can drive automatically and safely without conflict, reduce traffic congestion and improve the utilization ratio of road traffic. In this paper, we study the optimal path planning method based on improved Dijkstra algorithm, which provides an initial optimal path for each target. In order to solve the problem of increasing the time complexity and reducing the real-time performance of the classical Dijkstra algorithm, a dynamic path planning algorithm is proposed. It can dynamically add the required task nodes, eliminate redundant task nodes, so as to keep the total nodes of the road network unchanged, and then reduce the time complexity of the algorithm. The algorithm simulation software is programmed with VC, and the two algorithms are compared. The efficiency of the improved algorithm is proved. On the basis of cell objective path planning, a comprehensive path conflict solution based on the combination of global off-line planning and local online planning is proposed in this paper. When the multi-objective is in initial operation, the global planning is adopted. This paper provides an initial optimal running route for each target. In the course of operation, Petri net is used to build a conflict model for local planning of conflicting vehicles, and relevant test software is developed to verify the feasibility of the scheme. The hardware test platform is built and designed, and the road base station model is designed with STM32 microprocessor, and the corresponding wireless transceiver module is used to send and receive the data information from the model intelligent vehicle and process it. The feasibility of intelligent vehicle scheduling based on vehicle-road coordination is proved, which provides a reference for the development of intelligent transportation system in the future.
【学位授予单位】：大连理工大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：U495

【参考文献】