分心驾驶研究文献计量学分析

车辆自动化和信息娱乐系统的发展趋势使驾驶人分心成为日益突出的社会问题。为全面了解分心驾驶的研究进展，本文选取近 12 年关于分心驾驶的 2313 篇文章，利用文献计量工具VOSviewer和R-Bibliometrix系统梳理与分析分心驾驶的研究现状。首先，概述分心驾驶研究的总体情况，包括分心驾驶研究的国家分布、核心作者及核心期刊；其次，重点归纳分心驾驶领域的高影响力文献，并从“注意力机制”“分心驾驶风险与年轻驾驶人驾驶行为”“分心源”“分心驾驶检测”“自动驾驶下驾驶人的分心效应”这5个研究主题总结分心驾驶的研究热点；最后，构建分心驾驶的研究体系，并预测未来的发展趋势。研究认为：有必要加强研究分心的形成、消散和恢复机理；需扩展研究对象和场景，综合考虑车内和车外的分心源以及复合分心带来的风险；多源信息融合及考虑多类型分心可以进一步完善分心驾驶风险的研究；界定分心状态与等级划分，针对不同驾驶分心类型的识别应是未来分心驾驶检测研究的重点；自动驾驶场景下的分心行为及接管效能，以及混合交通流下的分心状态是值得关注的研究。基于机器视觉的分心检测和自动驾驶的接管效能将成为分心驾驶领域的研究前沿。

A Bibliometric Analysis of Distracted Driving Research

Trends in vehicle automation and infotainment systems have made driver distraction an increasingly prominent social problem. To understand the research progress of distracted driving comprehensively, this paper selects a total of 2313 articles on distracted driving in the past 12 years and systematically compares and analyzes the current status of distracted driving research using the bibliometric tools VOSviewer and R-Bibliometrix. First, the general situation of distracted driving research is outlined, including the country distribution, core authors and core journals of distracted driving research. Then, the high-impact literature in the field of distracted driving, and the hot topics of distracted driving research are summarized in five topics: "attention mechanism", "distracted driving risk and young drivers' driving behavior", "distraction source", "distracted driving detection" and "distraction effect of drivers under autonomous driving". At last, the research system of distracted driving is constructed, and the future development trend is predicted. The study concludes that: it is necessary to strengthen the research on the mechanism of distraction formation, dissipation and recovery. It is necessary to expand the research objects and scenarios to consider the sources of distraction inside and outside the vehicle and the risks caused by compound distraction. The fusion of multi-source information and the consideration of multiple types of distraction can further improve the research on distracted driving risks. The definition of distraction status and classification, and the identification of different types of driving distraction should be the focus of distracted driving detection in the future. The distracted behavior and takeover effectiveness in automatic driving scenarios and distraction states under mixed traffic flow are worthy of attention. Distraction detection based on computer vision and takeover effectiveness of automated driving will become the frontier of research in distracted driving.