Development of Evaluation Method for Vehicle Body Deformation in Accidents Using Polygon Mesh Model by Image Analysis

2020 Fiscal Year Final Research Report

• Project/Area Number: 18K13961
• Research Category: Grant-in-Aid for Early-Career Scientists
• Allocation Type: Multi-year Fund
• Review Section: Basic Section 25020:Safety engineering-related
• Research Institution: Suwa University of Science
• Principal Investigator: Kuniyuki Hiroshi 公立諏訪東京理科大学, 工学部, 教授 (30804223)
• Project Period (FY): 2018-04-01 – 2021-03-31
• Keywords: 交通事故分析 / 車体変形計測 / 乗員傷害予測

Outline of Final Research Achievements

This study can build a highly reproducible 3D vehicle body model using the SfM method by taking pictures with a high lap rate (approximately 95% or more) from a high point (approximately 7 m or more). It was verified that the amount of deformation of the deformed vehicle body can be evaluated at approximately 5% (± 10 mm).
Moreover, analysis using the US accident database clarified the effect of roof deformation for occupant injuries in side impact collision. Based on the results, this study constructed an occupant injury prediction model that takes into account the amount of vehicle body deformation, and improved the prediction sensitivity to approximately 88% compared to the conventional prediction model.

Free Research Field

自動車衝突安全・予防安全／交通事故分析

Academic Significance and Societal Importance of the Research Achievements

先進衝突自動通報（AACN）の適用拡大に向けて、重症度予測によるトリアージは最も重要な課題の一つであり、乗員傷害予測精度のさらなる向上が求められている。本研究の成果によって乗員傷害予測に対する車体変形量の影響が明らかになり、写真撮影から車体変形量を計測、評価する基盤が構築できた。さらに、得られた車体変形量を用いた乗員傷害予測式によって、事故現場でのトリアージの判断や乗員傷害予測精度の向上に貢献できると考える。