自動運転車がまぶしさによる事故リスクを低減(Self-driving cars can reduce dangers of sun glare on highways)

2025-07-31 コンコルディア大学

コンコルディア大学の研究は、日の出や日の入り時に起こる太陽光のまぶしさ(サングレア)が高速道路で追突事故などを引き起こす問題に対し、自動運転車(AV)の導入効果をシミュレーションで検証した。モントリオール近郊の高速道路をモデルにした結果、高度な自律性を持つ車両ほど速度調整や車間維持が安定し、急ブレーキや危険状況の発生が大幅に減少。完全自動化でなくても、一部の車両がAV化されるだけで交通全体の流れが改善し、視界不良下での事故リスクを低減できることが示された。研究チームは、自動運転技術を交通計画や政策に積極的に組み込むべきだと提言している。本成果は学術誌『Future Transportation』に掲載された。

<関連情報>

日射による眩光下での高速道路における自律走行車の安全対策の評価 Evaluating Autonomous Vehicle Safety Countermeasures in Freeways Under Sun Glare

Hamed Esmaeeli,Arash Mazaheri,Tahoura Mohammadi Ghohaki and Ciprian Alecsandru
Future Transportation  Published: 14 February 2025
DOI:https://doi.org/10.3390/futuretransp5010020

自動運転車がまぶしさによる事故リスクを低減(Self-driving cars can reduce dangers of sun glare on highways)

Abstract

The use of traffic simulation to analyze traffic safety and performance has become common in transportation engineering. Microsimulation methods are increasingly used to analyze driving performance for different road geometries and environmental elements. Drivers’ perception has an important impact on driving performance factors contributing to traffic safety on transportation facilities (highways, arterials, intersections, etc.). Impaired vision leads to failure in drivers’ perception and making right decisions. Various studies investigated the impact of environmental elements (fog, rain, snow, etc.) on driving performance. However, there is limited research examining the potentially detrimental effects on driving capabilities due to differing exposure to natural light brightness, in particular sun exposure. Autonomous vehicles (AVs) showed a significant impact enhancing traffic capacity and improving safety margins in car-following models. AVs may also enhance and/or complement human driving under deteriorated driving conditions such as sun glare. This study uses a calibrated traffic simulation and surrogate safety assessment model to improve traffic operations and safety performance under impaired visibility using different types of autonomous vehicles. A combination of visibility reduction, traffic flow characteristics, and autonomy levels of AVs was simulated and assessed in terms of the number of conflicts, severity level, and traffic operations. The simulation analysis results used to reveal the contribution of conflicts to the risk of crashes varied based on the influence of autonomy level on safe driving during sun glare exposure. The outcome of this study indicates the benefits of using different levels of AVs as a solution to driving under vision impairment situations that researchers, traffic engineers, and policy makers can use to enhance traffic operation and road safety in urban areas.

0108交通物流機械及び建設機械
ad
ad
Follow
ad
タイトルとURLをコピーしました