2018 年 12 月 Chinese Journal on Internet of Things December 2018
第 2 卷第 4 期 物 联 网 学 报 Vo l .2
No.4
矿山物联网人员情境描述与不安全行为识别
冯仕民
1,2,3
,刘忠育
1,2,3
,俞啸
1,2,3
,孟磊
1,2,3
,赵志凯
1,2,3
,丁恩杰
1,2,3
(1. 矿山互联网应用技术国家地方联合工程实验室,江苏 徐州 221008;
2. 中国矿业大学物联网(感知矿山)研究中心,江苏 徐州 221008;
3. 徐州市物联网产业发展研究中心,江苏 徐州 221008)
摘 要:矿山的环境复杂,智能化识别此类行为需要数据驱动方法和基于机器可读的领域知识的方法,然而矿山物
联网的数据缺乏语义信息,矿山人员状态信息和不安全行为知识没有标准的表示方法。为解决上述问题,提出了一
种基于语义本体的人员状态信息感知描述方法及一种数据驱动和知识驱动相结合的矿山人员不安全行为识别框架。首
先介绍了语义本体在物联网领域的概况,阐述了矿山人员不安全行为识别的重要性和必要性,介绍了人员行为识别、
情境感知和语义本体等研究背景,引出了矿山人员状态信息感知描述方法。基于此情境建模,提出了一种矿山人员不
安全行为识别框架并应用于矿山人员不佩戴防护与安全装备的不安全行为识别,最后总结并展望了人工智能方法在矿
山物联网应用层研究中的前景。
关键词:矿山物联网;不安全行为;行为识别;本体;人工智能
中图分类号:TP181
文献标识码:A
doi: 10.11959/j.issn.2096-3750.2018.00083
Workers’ context description and unsafe behavior
recognition in Internet of things for mines
FENG Shimin
1,2,3
, LIU Zhongyu
1,2,3
, YU Xiao
1,2,3,
MENG Lei
1,2,3
, ZHAO Zhikai
1,2,3
, DING Enjie
1,2,3
1. The National Joint Engineering Laboratory of Internet Applied Technology of Mines, Xuzhou 221008, China
2. Internet of Things (Perception Mine) Research Center, China University of Mining and Technology, Xuzhou 221008, China
3. The Xuzhou Research Center of Industry Development for Internet of Things, Xuzhou 221008, China
Abstract: The mine is a complicated environment. The intelligent recognition of such behavior requires not only the da-
ta-driven activity recognition method, but also the machine-readable domain knowledge based approach. However, the data
of IoT for mines lacks the semantic information. Besides, there is no standard way of describing the worker’s context and
representing the knowledge of worker’s unsafe behavior. In order to solve the above problems, a semantic ontology based
approach to describing the worker’s context and a hybrid method based framework for worker’s unsafe behavior recognition
were presented. This method combines the data-driven approach and the knowledge-driven approach. Firstly, an introduction
to the semantic ontology in the Internet of things was given. Then, the importance and necessity of worker’s unsafe behavior
recognition was introduced. After that, the research background on human activity recognition, context-awareness and se-
mantic ontology was presented. This was followed by the semantic ontology based approach to the worker’s context descrip-
tion. Based on the context modeling, the framework that combined the data-driven method and the knowledge-driven meth-
od for the worker’s unsafe behavior recognition was proposed. The application of the framework was illustrated with the
recognition of a kind of worker’s unsafe behavior who don’t wear the protective and safety equipment. Finally, the conclusions
were drawn and the prospect of using the artificial intelligence method in the application layer of mine IoT was presented.
Key words: mine Internet of things, unsafe behavior, activity recognition, ontology, artificial intelligence
收稿日期:2018-08-09;修回日期:2018-10-29
基金项目:国家重点研发计划项目“矿山安全生产物联网关键技术与装备研发”课题资助项目(No.2017YFC0804401)
Foundation Item: National Key Research and Development Program Project “Research and Development of Key Technologies an
Equipment for Internet of Things in Mine Safety Production” (No.2017YFC0804401)