Large-scale experiment on real-time imputation method for sparse floating car data

A floating car system is an effective way to collect traffic information without roadside sensors. However, because the number of floating cars that can feasibly be used is limited, there is a large amount of missing data with floating car data (FCD). In an effort to address this problem, an imputation method based on feature space projection is proposed. Since imputation is achieved on each feature space axis that represents the spatially correlated component of FCD, the method performs an accurate estimation based on sparse observations. This paper reports an evaluation result of the imputation method in a large-scale experiment with about 2,000 taxis in central Tokyo. For the covering abstract see ITRD E140665.

• Authors:
  • HIRUTA, T
  • KUMAGAI, M
  • SUZUKI, K
  • YOKOTA, T
• Publication Date: 2007

Language

• English

Media Info

• Serial:
  • PROCEEDINGS OF THE 14TH WORLD CONGRESS ON INTELLIGENT TRANSPORT SYSTEMS (ITS), HELD BEIJING, OCTOBER 2007
  • Publisher: ITS America

Subject/Index Terms

• TRT Terms: Conferences; Data collection; Information processing; Taxicabs; Traffic flow; Urban areas; Vehicles
• Geographic Terms: Japan
• Subject Areas: Public Transportation; Vehicles and Equipment; I72: Traffic and Transport Planning; I73: Traffic Control;

Filing Info

• Accession Number: 01163465
• Record Type: Publication
• Source Agency: Transport Research Laboratory
• Files: ITRD
• Created Date: Jul 22 2010 11:01AM