This Car is Mine!: Driver Pattern Dataset extracted from CAN-bus

We propose a driver pattern dataset consists of 51 features extracted from CAN (Controller Area Network) of Hyundai YF Sonata while four drivers drove city roads of Seoul, Republic of Korea. Under the belief that different driving patterns implicitly exist at CAN data, we collected CAN diagnosis data from four drivers in pursuit of research on driver identification, driver profiling, and abnormal driving behavior detection. Four drivers are named A, B, C, and D. The driver A performed eight trips, the driver B performed eight trips, the driver C performed five trips, and the driver D performed nine trips. We collected 51 features from CAN utilizing On-Board Diagnostic 2 (OBD-II) and CarbigsP as a scanning tool to extract features. Every feature is recorded by 1 second along with the trip. Note that the time consumed by each trip is all different as the traffic environment were all different during the data collection. In our research work 'This Car is Mine!: Automobile Theft Countermeasure Leveraging Driver Identification with Generative Adversarial Networks', we showed the dataset could be fully utilized to analyze the unique characteristics of each driver. We expect this driver pattern dataset to be a concrete baseline dataset for future studies on novel driving pattern recognition approaches.

本研究构建一款驾驶模式数据集，该数据集包含从现代YF索纳塔（Hyundai YF Sonata）的控制器局域网（Controller Area Network, CAN）中提取的51项特征，数据采集场景为四名驾驶员在韩国首尔城市道路的行驶过程。基于“不同驾驶员的驾驶模式会在CAN数据中隐性体现”这一核心假设，我们面向驾驶员身份识别、驾驶员画像构建以及异常驾驶行为检测三类研究目标，从四名驾驶员处采集了CAN诊断数据。本次研究涉及的四名驾驶员分别命名为A、B、C、D：驾驶员A完成8次行驶行程，驾驶员B完成8次，驾驶员C完成5次，驾驶员D完成9次。本次采集通过车载诊断二代（On-Board Diagnostic 2, OBD-II）设备与CarbigsP扫描工具从CAN总线中提取了共计51项特征。所有特征均以1秒为采样间隔随行驶行程同步记录。需说明的是，由于数据采集期间各行程遭遇的交通环境各不相同，因此单次行程的耗时存在差异。在我们的研究论文《This Car is Mine!: Automobile Theft Countermeasure Leveraging Driver Identification with Generative Adversarial Networks》中，我们已验证该数据集可被充分用于分析每位驾驶员的独特驾驶特征。我们期望该驾驶模式数据集能够为后续面向新型驾驶模式识别方法的研究提供可靠的基准数据集。

搜集汇总

背景与挑战

背景概述

该数据集是一个驾驶员模式数据集，包含从现代YF索纳塔汽车的CAN总线提取的51个特征，数据来自四名驾驶员在首尔城市道路上的驾驶行程，每秒记录一次。它旨在支持驾驶员识别、分析和异常行为检测研究，并已在实际研究中验证了其应用价值。