A fine-grained OD estimating system fusing sensors in buses

2023 Fiscal Year Final Research Report

• Project/Area Number: 20K11789
• Research Category: Grant-in-Aid for Scientific Research (C)
• Allocation Type: Multi-year Fund
• Section: 一般
• Review Section: Basic Section 60060:Information network-related
• Research Institution: Nara Institute of Science and Technology
• Principal Investigator: Arai Ismail 奈良先端科学技術大学院大学, 総合情報基盤センター, 准教授 (60512572)
• Project Period (FY): 2020-04-01 – 2024-03-31
• Keywords: ITS

Outline of Final Research Achievements

To achieve high-precision acquisition of origin-destination (OD) data for bus passengers, we evaluated the potential for generating OD data using each of the following information sources: WLAN or Bluetooth devices possessed by passengers, and in-vehicle camera footage. We aimed to enhance the accuracy of OD data generation by integrating these sources and considering indirect information such as weather conditions.
As a result, we obtained evaluation results for OD data generation using Bluetooth and camera footage as single information sources. However, during the research period, we did not achieve a solution for integrating multiple information sources, leaving this as a remaining challenge. The foundational validation indicated that camera footage was dominant, revealing the need for improvements in the acquisition methods of other single information sources.

Free Research Field

情報ネットワーク

Academic Significance and Societal Importance of the Research Achievements

本研究の学術的な成果は乗降客数の推定においては機械学習モデルの新たな考案や複数情報源による特徴量の策定によって高精度化を実現したことにある。ODデータ生成については複数情報源の活用手法の確立までには至らなかったが、常時取得可能であることから本研究成果の再現率が低く見えても、そもそもパーソントリップ調査に比べて飛躍的に量が増えるため、実用性向上に貢献できる。