In a groundbreaking study published in ‘Taiyuan Ligong Daxue xuebao’, researchers from Xi’an University of Science and Technology have unveiled a novel approach to accurately locate personnel working underground in coal mines. This development is crucial for enhancing safety measures in an industry where the risks are significant and often unpredictable.
The lead author, SHAO Xiaoqiang, highlights the importance of this research, stating, “The precise location of underground personnel is not just a technical challenge; it is a matter of life and death.” This sentiment underscores the urgency of addressing the challenges posed by non-line-of-sight (NLOS) interference, which can severely compromise the accuracy of ultra-wideband (UWB) positioning systems.
In their innovative method, the researchers combine pedestrian dead reckoning (PDR) and map information to filter out infeasible positions. They utilize multi-granularity mesh filters to estimate locations and headings while generating weak labels from the map data. This self-training approach significantly reduces the time and costs associated with traditional supervised learning methods, which often require extensive training data and manual labeling.
The study showcases impressive results, demonstrating that the root-mean-square error for NLOS scenarios decreased from 1.02 meters to just 0.32 meters. Furthermore, the research indicates a 69% improvement in ranging error, with the rate of positioning accuracy—defined as errors less than 0.3 meters—jumping from 49% to 89%. These advancements not only enhance operational safety but also have the potential to revolutionize how construction and mining companies manage their workforce underground.
As industries increasingly turn to automation and data-driven solutions, the implications of this research extend beyond coal mining. The methodologies developed could be adapted for various sectors where personnel safety is paramount, including construction, oil and gas, and emergency response. The ability to autonomously collect training data and improve positioning accuracy could lead to smarter, safer work environments.
Shao’s team has set a new benchmark in the quest for reliable underground positioning, paving the way for future innovations that could further integrate technology into the construction sector. As they continue to refine their methods, the research community and industry stakeholders alike will be closely watching how these findings can be implemented in real-world scenarios.
For more information about this transformative research, you can visit the College of Electrical and Control Engineering, Xi’an University of Science and Technology, where SHAO Xiaoqiang and his team are leading the charge in enhancing safety through technology.
