IPL: Image-Assisted Person Localization for Underground Coal Mines

Area of Science:

• Mining Engineering
• Computer Science
• Robotics

Background:

• Accurate underground personnel localization is critical for coal mine safety and operational efficiency.
• Existing localization methods struggle with cumulative errors in challenging underground environments.

Purpose of the Study:

• To develop a novel localization scheme for underground coal mines that overcomes the limitations of traditional methods.
• To improve the precision and reliability of personnel tracking in dark, subterranean environments.

Main Methods:

• Integration of step detection algorithms for initial positioning.
• Implementation of a specialized base station with cameras for image recognition.
• Utilizing miner's lamps to project unique shapes for precise location identification via image analysis.

Main Results:

• The proposed system effectively corrects cumulative errors inherent in step-based localization.
• Experimental validation in indoor and underground settings demonstrated high accuracy.
• Achieved 96% of spatial errors at 2.5 meters or less.

Conclusions:

• The novel localization scheme offers a robust solution for accurate underground personnel tracking.
• The combination of step detection and image recognition provides a reliable method for enhancing mine safety.