Proposal of IoT instruction system to evacuate diverse crowds from indoor facility in unstable situation

2021 Fiscal Year Final Research Report

• Project/Area Number: 18K13973
• Research Category: Grant-in-Aid for Early-Career Scientists
• Allocation Type: Multi-year Fund
• Review Section: Basic Section 25030:Disaster prevention engineering-related
• Research Institution: National Institute of Technology, Toyama College
• Principal Investigator: Ito Nao 富山高等専門学校, その他部局等, 講師 (30635214)
• Project Period (FY): 2018-04-01 – 2022-03-31
• Keywords: マルチエージェント / Social Force / IoT / 避難誘導 / 屋内避難 / 障害物

Outline of Final Research Achievements

In this study, we suggested a new crowd walking model in consideration of personal space. In this model, we reproduced the obstacle setting characteristic of pedestrian flow in the real experiment. By using this model, in virtual space which imitates a commercial facility, we evaluated the effectiveness of IoT indoor evacuation instruction system which we proposed. As a result, it is suggested that there are dynamic optimal solutions in operation policy of the system such as the instruction interval or the priority to avoid crowds. Furthermore, we performed trail experiments for the proposed system and confirmed that it worked effectively. From the above, we can say that we show the effectiveness and feasibility of our proposed system.

Free Research Field

マルチエージェント

Academic Significance and Societal Importance of the Research Achievements

本研究の学術的意義は，①パーソナルスペースを考慮した群集歩行モデルを提案し被験者実験において観察される障害物設置特性を再現したこと，②IoT屋内避難誘導システムの運用方法における動的最適解の存在を示唆したことの2点である．①により，障害物を複数個設置した場合の最適設置位置や商業施設における店内レイアウトの最適化への応用が期待できる．②について，これまでに類似のシステムは既にいくつか提案されいたものの最短経路による誘導や静的なパラメータでの運用が主流だった．これに対し，常に変化する避難状況に応じて内部パラメータを動的に変化させ，より安全で迅速な避難が可能であることを示唆した本研究の意義は大きい．