Affiliation:
1. Purdue University, 465 Northwestern Avenue, West Lafayette, IN 47907-2035.
Abstract
The traffic industry has invested in an extensive infrastructure for vehicle detection and speed estimation. This paper describes how these sensors can be enhanced to measure not only the presence and speed of a vehicle but also its acceleration and travel time. While acceleration of vehicles is of independent interest, knowledge of the acceleration of a vehicle can also be used to normalize vehicular signatures to increase significantly the probability of successfully reidentifying vehicles to generate travel time estimates. In this paper, the acceleration correction for vehicle reidentification is developed and analyzed in terms of the percentage of vehicles correctly matched and in the framework of a communication system's receiver operating characteristic. The percentage of vehicles correctly matched for the travel time estimation is increased from 55.1% when only the speed of the passing vehicles is used to 64.5% and the acceleration estimates are also used.
Subject
Mechanical Engineering,Civil and Structural Engineering
Cited by
2 articles.
订阅此论文施引文献
订阅此论文施引文献,注册后可以免费订阅5篇论文的施引文献,订阅后可以查看论文全部施引文献