Research Article
Key Location Discovery of Underground Personnel Trajectory Based on Edge Computing
@INPROCEEDINGS{10.1007/978-3-030-64214-3_9, author={Zhao Jinjin and Zou Xiangyu and Zhang Yu and Gu Youya and Wu Fan and Zhu Zongwei}, title={Key Location Discovery of Underground Personnel Trajectory Based on Edge Computing}, proceedings={Mobile Computing, Applications, and Services. 11th EAI International Conference, MobiCASE 2020, Shanghai, China, September 12, 2020, Proceedings}, proceedings_a={MOBICASE}, year={2020}, month={12}, keywords={Personnel trajectories analysis Inflection point Stay point Edge computing}, doi={10.1007/978-3-030-64214-3_9} }- Zhao Jinjin
Zou Xiangyu
Zhang Yu
Gu Youya
Wu Fan
Zhu Zongwei
Year: 2020
Key Location Discovery of Underground Personnel Trajectory Based on Edge Computing
MOBICASE
Springer
DOI: 10.1007/978-3-030-64214-3_9
Abstract
With the rapid development of smart mines, workers’ trajectories can be accurately tracked and generate massive positioning data. However, how to quickly find useful information in a large amount of data is an important issue at present. Consequently, in this paper, we propose an algorithm for application in underground edge computing systems, called key location discovery (KLD). First, the algorithm reconstructs the trajectory data by the potential semantic information of the underground environment and miners work types to be more suitable for the actual situation. Then, the KLD algorithm screen out the key locations of underground personnel trajectories according to inflection point and stay point. In the meanwhile, compared with the trajectory structure-based hot spots (TS_HS) discovery algorithm, KLD algorithm reduced the positioning data by 1/4 and calculating time. In addition, placing the algorithm proposed in this paper on the edge side for calculation and processing can filter out key information in real time, which is more beneficial to the follow-up work, including the study of personnel trajectory abnormalities and prediction.
