A Filter Design Based on Human Sentiments Fusion for Estimating Vehicle Arrival Time

Over the past few years, several algorithms have been developed to predict travel time. One of the most important algorithms is Kalman filter, which has been widely used in estimating the state of the traffic. However, Kalman filter usually uses sensory data that are collected from sensors such as global positioning system (GPS). These devices have the disadvantages of requiring sophisticated facilities, failure if its circuitry is declined and inconstant accuracy depending on the signaling strength. These devices are very expensive and may be not available. In this paper, we propose a novel model that can be considered as a fusion filter in order to predict the travel time in accurate manner based on DSmT (Dezert-Smarandache Theory) as a fusion technique. Thus, fusion theories and data combination relationships are exploited to construct such transportation filter. That filter takes human sentiments as data source, computes the pignistic probability of the traffic status, accordingly, generates the expected vehicle arrival time. As a case study, we applied this filter on the data collected from Cairo, which is one of the most congested cities all over the world. The fusion process is performed using people reports from the Internet that replace the sensory data. Moreover, we enhanced our filter using empirical sentimental formulas to obtain more accurate results compared with Kalman filter.

  • Record URL:
  • Availability:
  • Supplemental Notes:
    • © Springer Science+Business Media, LLC, part of Springer Nature 2018. The contents of this paper reflect the views of the authors and do not necessarily reflect the official views or policies of the Transportation Research Board or the National Academy of Sciences.
  • Authors:
    • Hisham, Basma
    • Hamouda, Alaa
    • Zaki, Mohamed
  • Publication Date: 2018-9

Language

  • English

Media Info

Subject/Index Terms

Filing Info

  • Accession Number: 01680656
  • Record Type: Publication
  • Files: TRIS
  • Created Date: Aug 7 2018 8:39AM