SafeDrive: Online Driving Anomaly Detection From Large-Scale Vehicle Data
SafeDrive: Online Driving Anomaly Detection From Large-Scale Vehicle Data
复制标题
SafeDrive：大规模车辆数据在线驾驶异常检测
DOI：
10.1109/tii.2017.2674661
复制
发表时间：
2017-02
影响因子：
12.3
通讯作者：
Lin Xuelian
中科院分区：
文献类型：
--
作者：
Zhang Mingming;Chen Chao;Wo Tianyu;Xie Tao;Bhuiyan Md Zakirul Alam;Lin Xuelian
识别驾驶异常对于提高行车安全具有重要意义。随着车联网（IoV）技术的发展，从多个车辆传感器获取大数据成为可能，这些大数据在识别驾驶异常方面发挥着基础性作用。现有的方法主要是基于规则或监督学习。然而，这种方法通常需要标记数据，这在大数据场景中通常不可用。此外，由于驾驶行为在车辆状态下不同（例如，速度和档位），为了精确地对驾驶行为进行建模，需要融合多个传感器数据源。为了解决这些问题，在本文中，我们提出了SafeDrive，一种在线和状态感知的方法，它不需要标记的数据。从历史数据集，SafeDrive统计离线导出状态图（SG）作为行为模型。然后，SafeDrive将在线数据流分割成段，并将每个段与SG进行比较。SafeDrive将显著偏离SG的段识别为异常。我们在基于云的车联网平台上评估SafeDrive，该平台拥有超过29，000辆真实的联网车辆。评估结果表明，SafeDrive能够从大规模车辆数据流中有效识别各种驾驶异常，总体准确率为93%;这些识别出的驾驶异常可以用于及时提醒驾驶员纠正其驾驶行为。
Identifying driving anomalies is of great significance for improving driving safety. The development of the Internet-of-Vehicle (IoV) technology has made it feasible to acquire big data from multiple vehicle sensors, and such big data play a fundamental role in identifying driving anomalies. Existing approaches are mainly based on either rules or supervised learning. However, such approaches often require labeled data, which are typically not available in big data scenarios. In addition, because driving behaviors differ under vehicle statuses (e.g., speed and gear position), to precisely model driving behaviors needs to fuse multiple sources of sensor data. To address these issues, in this paper, we propose SafeDrive, an online and status-aware approach, which does not require labeled data. From a historical dataset, SafeDrive statistically offline derives a state graph (SG) as a behavior model. Then, SafeDrive splits the online data stream into segments and compares each segment with the SG. SafeDrive identifies a segment that significantly deviates from the SG as an anomaly. We evaluate SafeDrive on a cloud-based IoV platform with over 29 000 real connected vehicles. The evaluation results demonstrate that SafeDrive is capable of identifying a variety of driving anomalies effectively from a large-scale vehicle data stream with an overall accuracy of 93%; such identified driving anomalies can be used to timely alert drivers to correct their driving behaviors.
登录
查看更多内容
DOI：
10.1109/iscc.2015.7405502
发表时间：
2015-07
期刊：
2015 IEEE Symposium on Computers and Communication (ISCC)
影响因子：
--
作者：
A. Taha;N. Nasser
通讯作者：
A. Taha;N. Nasser
DOI：
10.1109/icsenst.2013.6727671
发表时间：
2013-12
期刊：
2013 Seventh International Conference on Sensing Technology (ICST)
影响因子：
--
作者：
T. Chakravarty;Avik Ghose;Chirabrata Bhaumik;Arijit Chowdhury
通讯作者：
T. Chakravarty;Avik Ghose;Chirabrata Bhaumik;Arijit Chowdhury
DOI：
10.1109/tits.2012.2187640
发表时间：
2012-09-01
影响因子：
8.5
作者：
Fazeen, Mohamed;Gozick, Brandon;Gonzalez, Marta C.
通讯作者：
Gonzalez, Marta C.
DOI：
--
发表时间：
2014
期刊：
--
影响因子：
--
作者：
Diana María;Gómez Jaramillo;Claudia Victoria;Isaza Narváez
通讯作者：
Diana María;Gómez Jaramillo;Claudia Victoria;Isaza Narváez
DOI：
10.1145/2632048.2632077
发表时间：
2013-12
期刊：
Proceedings of the 2014 ACM International Joint Conference on Pervasive and Ubiquitous Computing
影响因子：
--
作者：
Bernhard Firner;Shridatt Sugrim;Yulong Yang;J. Lindqvist
通讯作者：
Bernhard Firner;Shridatt Sugrim;Yulong Yang;J. Lindqvist