Quantifying the Measurement Precision of a Commercial Ultrasonic Real-Time Location System for Camera Pose Estimation...

Area of Science:

• Geomatics Engineering
• Robotics and Automation
• Computer Vision

Background:

• Indoor photogrammetry often requires manual ground control points (GCPs) or integrated sensors.
• Global Navigation Satellite System (GNSS) technology provides direct georeferencing for aerial photogrammetry, but lacks indoor equivalents.
• Establishing the measurement precision of indoor positioning systems is essential before their photogrammetric application.

Purpose of the Study:

• To characterize the repeatability and temporal stability of the ZeroKey Quantum real-time location system (RTLS).
• To establish the measurement precision of an ultrasonic RTLS as a prerequisite for photogrammetric bundle adjustment.
• To determine optimal data collection protocols for indoor positioning systems.

Main Methods:

• Systematic tripod-mounted observations at 30 indoor test locations.
• Development of an automated stationary detection algorithm (20 mm threshold).
• Quantification of measurement precision through analysis of position and orientation uncertainties.

Main Results:

• An automated stationary detection algorithm identified all stationary periods within 30 seconds.
• Optimal 1-second observation windows achieved 3 mm position and 1° orientation precision.
• Quantified per-axis root mean square (RMS) uncertainties of 1.6 mm, 1.7 mm, 1.1 mm for position and 0.08°, 0.09°, 0.07° for orientation.

Conclusions:

• Ultrasonic RTLS demonstrates millimeter-level position repeatability and sub-degree orientation repeatability.
• The established measurement precision is sufficient for subsequent accuracy testing in photogrammetric bundle adjustment.
• This research paves the way for precise indoor 3D reconstruction without labor-intensive GCPs.