• Search Research Projects
  • Search Researchers
  • How to Use
  1. Back to previous page

Development of Human-in-the-Loop Human Mobility Simulation

Research Project

Project/Area Number 21K14260
Research Category

Grant-in-Aid for Early-Career Scientists

Allocation TypeMulti-year Fund
Review Section Basic Section 22050:Civil engineering plan and transportation engineering-related
Research InstitutionThe University of Tokyo

Principal Investigator

Pang Yanbo  東京大学, 空間情報科学研究センター, 特任助教 (60870178)

Project Period (FY) 2021-04-01 – 2024-03-31
Project Status Completed (Fiscal Year 2023)
Budget Amount *help
¥4,680,000 (Direct Cost: ¥3,600,000、Indirect Cost: ¥1,080,000)
Fiscal Year 2023: ¥780,000 (Direct Cost: ¥600,000、Indirect Cost: ¥180,000)
Fiscal Year 2022: ¥1,560,000 (Direct Cost: ¥1,200,000、Indirect Cost: ¥360,000)
Fiscal Year 2021: ¥2,340,000 (Direct Cost: ¥1,800,000、Indirect Cost: ¥540,000)
Keywords人の流れ / エージェントシミュレーション / データ基盤 / 人間参加型機械学習 / 生成系AI / 擬似人流 / 深層学習 / 強化学習 / 対話型機械学習
Outline of Research at the Start

近年、数多く人流シミュレーションに関する研究が行われてきたが、行動モデルの訓練や結果の評価には、依然、高価かつ入手しにくい携帯電話データに依存し、 事前にデータが入手できない未曾有の場面に対して、高精度な人流を再現・予測できない問題が残っている。そこで、本研究では、シミュレーションモデルと人間の知能を組み合わせて、シミュレーションの再現性と信頼性を向上するために仕組みを提案 する。そこでは、人間の評価をエージェントモデルへフィードバックすることで、シミュレ ーション精度の向上を図る。 この成果は、これまで不可能だった状況の人流を再現することを可能とし、社会インフラデータの整備に貢献する。

Outline of Final Research Achievements

In this study, we combined simulation models and human intelligence to improve the reproducibility and reliability of simulations. The main achievements are as follows: 1. Creation and provision of nationwide pseudo-people mobility data: Using agent-based modeling and multi-source open data, we created and provided a dataset that replicates a typical weekday people flow across Japan. 2. Development of the generative AI model "MobilityGPT": We developed a generative AI model based on large-scale mobility data, contributing to urban planning and traffic management.
These achievements aid in the design of efficient transportation networks and quick evacuation planning during disasters.

Academic Significance and Societal Importance of the Research Achievements

学術的意義:本研究は、シミュレーションモデルと人間の知能を融合させた新手法により、エージェントベースモデリングと生成系AIモデルの精度向上を実現した。特に、生成系AIモデル「MobilityGPT」は大規模な移動データに基づき、複雑な人間の行動パターンの再現を可能にし、都市情報学や交通工学に新たな知見を提供した。
社会的意義:全国擬似人流データの提供により、詳細な移動パターンの情報が得られ、政策の感度分析や商業施設の配置計画、大規模災害時の避難シミュレーションなどの研究や施策に活用可能となった。デジタルツインやスマートシティにも寄与し、社会全体の持続可能性向上に貢献する。

Report

(4 results)
  • 2023 Annual Research Report   Final Research Report ( PDF )
  • 2022 Research-status Report
  • 2021 Research-status Report
  • Research Products

    (14 results)

All 2023 2022 2021

All Journal Article (4 results) (of which Int'l Joint Research: 1 results,  Peer Reviewed: 4 results) Presentation (7 results) (of which Int'l Joint Research: 3 results) Funded Workshop (3 results)

  • [Journal Article] Deep Learning for Destination Choice Modeling: A Fundamental Approach for National Level People Flow Reconstruction2022

    • Author(s)
      Pang Yanbo、Sekimoto Yoshihide
    • Journal Title

      2022 IEEE International Conference on Big Data (Big Data)

      Volume: - Pages: 1900-1905

    • DOI

      10.1109/bigdata55660.2022.10020165

    • Related Report
      2022 Research-status Report
    • Peer Reviewed
  • [Journal Article] Spatial Attention Based Grid Representation Learning For Predicting Origin?Destination Flow2022

    • Author(s)
      Cai Mingfei、Pang Yanbo、Sekimoto Yoshihide
    • Journal Title

      2022 IEEE International Conference on Big Data (Big Data)

      Volume: - Pages: 485-494

    • DOI

      10.1109/bigdata55660.2022.10021023

    • Related Report
      2022 Research-status Report
    • Peer Reviewed
  • [Journal Article] Uncertainty of Traffic Congestion Estimation Using Nationwide Pseudo Trip Data and Agent-Based Simulation2022

    • Author(s)
      Tewari Aayush、Pang Yanbo、Sekimoto Yoshihide
    • Journal Title

      2022 IEEE International Conference on Big Data (Big Data)

      Volume: - Pages: 3854-3863

    • DOI

      10.1109/bigdata55660.2022.10020749

    • Related Report
      2022 Research-status Report
    • Peer Reviewed
  • [Journal Article] Simulating Human Mobility with Agent-based Modeling and Particle Filter Following Mobile Spatial Statistics2021

    • Author(s)
      Cai Mingfei、Pang Yanbo、Kashiyama Takehiro、Sekimoto Yoshihide
    • Journal Title

      Proceedings of the 29th International Conference on Advances in Geographic Information Systems

      Volume: - Pages: 411-414

    • DOI

      10.1145/3474717.3484203

    • Related Report
      2021 Research-status Report
    • Peer Reviewed / Int'l Joint Research
  • [Presentation] Deep Learning Approach to Logistics Trips Generation: Enhancing Pseudo People Flow with Agent-based Modeling2023

    • Author(s)
      Zhang, K., Pang, Y., and Sekimoto, Y
    • Organizer
      IEEE ITSC-2023
    • Related Report
      2023 Annual Research Report
    • Int'l Joint Research
  • [Presentation] Synthetic Network Traffic Data Generation using Deep Generative Models2023

    • Author(s)
      Yanbo Pang, Pierre Ferry, Kunyi Zhang
    • Organizer
      NetMob 2023 Book of Abstracts
    • Related Report
      2023 Annual Research Report
    • Int'l Joint Research
  • [Presentation] Towards Pseudo People Flow: Developing a Deep Generative Model based on PT Data to Reproduce Large-Scale Daily People Activity Profiles2023

    • Author(s)
      Yurong ZHANG, Kunyi ZHANG, Yanbo PANG, Yoshihide SEKIMOTO
    • Organizer
      第32回地理情報システム学会学術研究発表大会講演論文集
    • Related Report
      2023 Annual Research Report
  • [Presentation] 全国擬似人流データの提供と評価2022

    • Author(s)
      Yanbo Pang,樫山武浩,関本義秀,
    • Organizer
      第31回地理情報システム学会研究発表大会
    • Related Report
      2022 Research-status Report
  • [Presentation] 擬似人流データにおける時刻表を考慮した自治体全域の交通手段の推計 ―静岡県裾野市を対象に―2022

    • Author(s)
      笠原有貴,Yanbo Pang,樫山武浩,関本義秀,瀬崎薫
    • Organizer
      第31回地理情報システム学会研究発表大会
    • Related Report
      2022 Research-status Report
  • [Presentation] シナリオに基づく擬似人流生成のためのシミュレーション基盤の構築2022

    • Author(s)
      澁谷 遊野,Yanbo Pang,関本 義秀
    • Organizer
      第31回地理情報システム学会研究発表大会
    • Related Report
      2022 Research-status Report
  • [Presentation] Development of a Reinforcement Learning based Agent Model and People Flow Data to Mega Metropolitan Area2021

    • Author(s)
      Pang Yanbo、Kashiyama Takehiro、Sekimoto Yoshihide
    • Organizer
      IEEE International Conference on Big Data (Big Data)
    • Related Report
      2021 Research-status Report
    • Int'l Joint Research
  • [Funded Workshop] IEEE International Conference on Big Data2022

    • Related Report
      2022 Research-status Report
  • [Funded Workshop] IEEE International Conference on Big Data2021

    • Related Report
      2021 Research-status Report
  • [Funded Workshop] International Conference on Advances in Geographic Information Systems2021

    • Related Report
      2021 Research-status Report

URL: 

Published: 2021-04-28   Modified: 2025-01-30  

Information User Guide FAQ News Terms of Use Attribution of KAKENHI

Powered by NII kakenhi