• 主题
高级搜索  >
历史搜索 清空历史
首页> 外文学位 >Automated traffic incident detection using GPS-based transit probe vehicles.
【24h】

Automated traffic incident detection using GPS-based transit probe vehicles.

机译：使用基于GPS的运输探测车进行自动交通事件检测。

获取原文
获取原文并翻译 | 示例
页面导航
  • 摘要
  • 著录项
  • 相似文献
  • 相关主题

摘要

The recent advancements in electronics, communication and information processing and the application of such technologies to transportation through Intelligent Transportation Systems (ITS) have resulted in safer, faster and more efficient surface transportation systems. ITS have contributed significantly to the current state of technology in automated traffic monitoring, and probe vehicle-based traffic monitoring is one of these techniques. Although the potential of probe vehicle-based traffic monitoring was well understood, its wider application was severely restricted by the state of vehicle positioning system capabilities and associated cost considerations.; This thesis presents the devolvement and performance analysis of a probe vehicle-based automated traffic incident detection system that uses transit vehicles equipped with GPS as a passive probe vehicle fleet. Since these vehicles serve primary purposes other than providing vehicle-tracking data, such vehicles are more likely to be equipped with systems such as GPS, thus providing multiple benefits. However, passive probe vehicle-based traffic incident detection has its own challenges over dedicated probe vehicles. These challenges are addressed and solutions are developed in two stages in this thesis.; Firstly, the need for a GPS positioning technique that provides better performance in urban environments is addressed with an outline of advantages of High Sensitivity GPS (HSGPS). HSGPS error mitigation techniques are discussed with the emphasis on map-matching augmentations; a major advantage being their independence from vehicle sensors. A discussion of HSGPS map-matching issues, pros and cons of using internally filtered HSGPS positions, and alternative filtering techniques are also presented.; Secondly, probe-based incident detection algorithms and their performances are analyzed. Instead of conducting a complex field data collection for probe and traffic incident data collection, a traffic microsimulation model was developed to simulate transit probe vehicles with and without simulated incidents. The simulation model is calibrated using actual transit vehicle and non-transit vehicle data collected over several days. Incidents detection performance was analyzed with respect to three indicators: Detection Rate (DR), Time to Detect (TTD) and False Alarm Rate (FAR). Performance under varying incident characteristics and GPS performance levels were investigated and the results show over a 90% DR and a TTD of under 5 minutes for all simulated scenarios.
机译：电子，通信和信息处理方面的最新进展以及此类技术在通过智能运输系统（ITS）进行运输中的应用已导致更安全，更快和更高效的地面运输系统。 ITS极大地推动了自动交通监控技术的发展，而基于探测车的交通监控就是其中之一。尽管对基于探测车辆的交通监控的潜力已广为人知，但其广泛应用受到车辆定位系统功能状态和相关成本考虑的严重限制。本文介绍了基于探测车的自动交通事件检测系统的发展和性能分析，该系统使用配备GPS的过境车辆作为被动探测车队。由于这些车辆除了提供车辆跟踪数据外还用于其他主要目的，因此此类车辆更可能配备有GPS等系统，从而提供了多种好处。但是，基于被动探测车的交通事件检测相对于专用探测车具有其自身的挑战。本文解决了这些挑战，并在两个阶段中提出了解决方案。首先，以高灵敏度GPS（HSGPS）的优点概述了对在城市环境中提供更好性能的GPS定位技术的需求。讨论了HSGPS差错缓解技术，重点是地图匹配增强。其主要优点是它们独立于车辆传感器。还讨论了HSGPS地图匹配问题，使用内部过滤的HSGPS位置的优缺点以及其他过滤技术。其次，分析了基于探针的事件检测算法及其性能。代替进行复杂的现场数据收集以进行探测和交通事故数据收集，开发了交通微仿真模型来模拟有或没有模拟事故的过境探测车辆。使用实际的过境车辆和几天内收集的非过境车辆数据对仿真模型进行校准。针对三个指标分析了事件检测性能：检测率（DR），检测时间（TTD）和误报率（FAR）。调查了各种事件特征和GPS性能水平下的性能，结果显示所有模拟场景的DR均超过90％，TTD不到5分钟。

著录项

  • 作者

    Basnayake, Chaminda.;

  • 作者单位

    University of Calgary (Canada).;

  • 授予单位 University of Calgary (Canada).;
  • 学科 Engineering Civil.
  • 学位 Ph.D.
  • 年度 2004
  • 页码 270 p.
  • 总页数 270
  • 原文格式 PDF
  • 正文语种 eng
  • 中图分类 建筑科学;
  • 关键词

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱：kefu@zhangqiaokeyan.com

京公网安备：11010802029741号 ICP备案号：京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有

  • 客服微信

  • 服务号