||
本文为美国宾夕法尼亚州立大学(作者:Sneha Kadetotad)的硕士论文,共82页。
近年来,不使用GPS的道路车辆定位问题引起了人们的极大兴趣,并提出了许多解决方案。传统的车辆定位方法分为两类:基于特征向量匹配的全局定位和基于粒子滤波、卡尔曼滤波等技术的局部跟踪。这项工作提出了一种统一的方法,将基于特征的全局搜索鲁棒性与粒子滤波器的局部跟踪能力结合起来。本文利用宾夕法尼亚州I-80州际公路和220号美国公路沥青测量得到的特征向量,证明了车辆广域定位和局部跟踪的计算效率。
The localization of vehicles on roadwayswithout the use of a GPS has been of great interest in recent years and anumber of solutions have been proposed for the same. The localization ofvehicles has traditionally been divided by their solution approaches into twodifferent categories: global localization which uses feature-vector matching,and local tracking which has been dealt with using techniques like Particlefiltering or Kalman Filtering. This effort proposes a unifying approach thatcombines the feature-based robustness of global search with the local trackingcapabilities of a Particle filter. Using feature vectors produced from pitchmeasurements from Interstate I-80 and US Route 220 in Pennsylvania, this workdemonstrates wide area localization of a vehicle with the computationalefficiency of local tracking.
更多精彩文章请关注公众号:
Archiver|手机版|科学网 ( 京ICP备07017567号-12 )
GMT+8, 2024-10-20 04:33
Powered by ScienceNet.cn
Copyright © 2007- 中国科学报社