Evaluating Forward Collision Warning and Autonomous Emergency Braking Systems in India Using Dashboard Cameras

Author:

Shaikh Junaid1,Lubbe Nils2

Affiliation:

1. Autoliv India Pvt. Ltd.

2. Autoliv Development AB

Abstract

<div class="section abstract"><div class="htmlview paragraph">Road traffic fatalities in India have been increasing, reaching around 150,000 fatalities a year. To reduce fatalities, some prospective studies suggested using active safety technologies such as Forward Collision Warning (FCW), and Autonomous Emergency Braking (AEB). However, the effectiveness of FCW and AEB on Indian roads using retrospective studies is not known. Vehicle data such as radar, and controller area network signals could be used for the evaluation of the systems (FCW and AEB). However, these data are not readily accessible. This exploratory study aims to explore the opportunities and limitations of using simple dashboard cameras for a Field Operational Test. One European car with state-of-the-art FCW and AEB systems was rented. Fifteen drivers shared the vehicle, driving almost 10,000 km over 29 days. The vehicle was mounted with a set of dashboard cameras. The navigator noted the “system activated” events and “no activation” events in the logbook during the drive. Post completion of the driving activity, the system activated events: single event (only FCW) or combined event (FCW + AEB), and no activation events were analyzed. Three evaluators classified each system activated event as either a true positive or a false positive. Further, no activation events, where the driver felt the FCW should but did not activate, were classified as false negatives. A total of 79 single and combined events were identified. The AEB system produced 2.9 true positives and 0.2 false positives per 1000 km, while the FCW system produced 6.3 true positives and 2.1 false positives per 1000 km. For the FCW system, there were also 0.3 false negatives per 1000 km. The inter-rater reliability for the three evaluators was moderate indicating that not enough data is provided to accurately understand and classify the events. Reliable performance evaluation with the chosen simple approach seems highly challenging.</div></div>

Publisher

SAE International

Reference33 articles.

1. Ministry of Road Transport and Highways 2020

2. World Health Organization 2018

3. Mohan , D. , Tsimhoni , O. , Sivak , M. , and Flannagan , M.J. 2009

4. Mehta , V. , Babal , S. , Handa , M. et al. Road Safety in India – Joint Working Group Recommendations 3rd Global Ministerial Conference on Road Safety Stockholm, Sweden 2020

5. Sugimoto , Y. and Sauer , C. Effectiveness Estimation Method for Advanced Driver Assistance System and its Application to Collision Mitigation Brake System 19th International Technical Conference on the Enhanced Safety of Vehicles USA 2005

同舟云学术

1.学者识别学者识别

2.学术分析学术分析

3.人才评估人才评估

"同舟云学术"是以全球学者为主线,采集、加工和组织学术论文而形成的新型学术文献查询和分析系统,可以对全球学者进行文献检索和人才价值评估。用户可以通过关注某些学科领域的顶尖人物而持续追踪该领域的学科进展和研究前沿。经过近期的数据扩容,当前同舟云学术共收录了国内外主流学术期刊6万余种,收集的期刊论文及会议论文总量共计约1.5亿篇,并以每天添加12000余篇中外论文的速度递增。我们也可以为用户提供个性化、定制化的学者数据。欢迎来电咨询!咨询电话:010-8811{复制后删除}0370

www.globalauthorid.com

TOP

Copyright © 2019-2024 北京同舟云网络信息技术有限公司
京公网安备11010802033243号  京ICP备18003416号-3