Scholarlink

<p>Road safety remains a critical global concern, often requiring significant time and resources through traditional mitigation approaches. As a cost-effective and affordable solution to improve road safety, this study explores the potential of Advanced Driver Assistance Systems (ADAS), specifically the Collision Avoidance Systems (CAS) and Driver Monitoring Systems (DMS), towards road crash occurrences on a 150-km stretch of National Highway (NH)-65 in India, from Malkapur village to Kodad town. Alerts analyzed include Forward Collision Warning (FCW), Headway Monitoring Warning (HMW), Lane Departure Warning (LDW), Pedestrian Collision Warning (PCW), and inattention alerts such as Drowsiness, Asleep, and Distraction. The corridor was segmented into 500-m sections using Geographic Information System (GIS) tools, enabling the integration of ADAS alert data and road crash records. Statistical analyses were performed using IBM-SPSS tools, including Pearson correlation, Artificial Neural Network-Multilayer Perceptron (ANN-MLP) model, and Generalized Linear Modelling (GLM). Results indicate that HMW and LDW alerts are the most significant predictors of crash occurrences, reflecting unsafe driving behaviours like tailgating and abrupt lane changes. While GLM confirmed their strong linear relationship with crashes, ANN-MLP captured complex, nonlinear associations and highlighted the latent influence of drowsiness. To further check the impact of ADAS on crash occurrences, additional data from February and March 2024 showed a decline in both collision alerts and modelled crash severity over time, suggesting positive behavioural adaptation through ADAS alerts. The study concludes that ADAS, supported by driver training, offers a practical, scalable solution to enhance road safety by mitigating high-risk driving behaviours.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Impact of ADAS on Road Crashes: An Interurban Highway Case Study

  • 期刊论文

  • 发表日期：

    2025年10月18日

查看全文
  • Dev Singh Thakur,
  • Chalumuri Ravi Sekhar,
  • Anbumani Subramanian

摘要

Road safety remains a critical global concern, often requiring significant time and resources through traditional mitigation approaches. As a cost-effective and affordable solution to improve road safety, this study explores the potential of Advanced Driver Assistance Systems (ADAS), specifically the Collision Avoidance Systems (CAS) and Driver Monitoring Systems (DMS), towards road crash occurrences on a 150-km stretch of National Highway (NH)-65 in India, from Malkapur village to Kodad town. Alerts analyzed include Forward Collision Warning (FCW), Headway Monitoring Warning (HMW), Lane Departure Warning (LDW), Pedestrian Collision Warning (PCW), and inattention alerts such as Drowsiness, Asleep, and Distraction. The corridor was segmented into 500-m sections using Geographic Information System (GIS) tools, enabling the integration of ADAS alert data and road crash records. Statistical analyses were performed using IBM-SPSS tools, including Pearson correlation, Artificial Neural Network-Multilayer Perceptron (ANN-MLP) model, and Generalized Linear Modelling (GLM). Results indicate that HMW and LDW alerts are the most significant predictors of crash occurrences, reflecting unsafe driving behaviours like tailgating and abrupt lane changes. While GLM confirmed their strong linear relationship with crashes, ANN-MLP captured complex, nonlinear associations and highlighted the latent influence of drowsiness. To further check the impact of ADAS on crash occurrences, additional data from February and March 2024 showed a decline in both collision alerts and modelled crash severity over time, suggesting positive behavioural adaptation through ADAS alerts. The study concludes that ADAS, supported by driver training, offers a practical, scalable solution to enhance road safety by mitigating high-risk driving behaviours.