Answers are grounded in this paper and cite their sources.
This paper investigates the effects of ground-based visual beacon geometry, camera pose, and single-beacon loss on the accuracy-based height limits of UAV localization. Regular configurations with 3–6 peripheral beacons and variants with a central beacon are considered. Monte Carlo simulations of PnP/P3P localization cover beacon-field radii of 10–50 m, projection errors of 0.5–5.0 pixels, horizontal camera offsets up to 1.5 field radii, and optical-axis tilts up to 40°. The geometric height limit is determined using a mean 3D localization error criterion of no more than 1 m and a successful-trial rate of at least 0.99. At a 30 m radius and 2-pixel error, increasing peripheral beacons from four to six raises the height limit from 63.4 to 67.3 m. Peripheral-beacon loss reduces it by 12.0% for the square, 7.0% for the pentagon, and 3.8% for the hexagon with a central beacon. At an offset of 1.5 field radii and 40° tilt, the geometric height limit for configurations with 4–6 peripheral beacons and their central-beacon variants exceeds the nominal value by 28–35%, whereas the optical height limit is defined for only 17–21% of directions. Power-law approximations with coefficients of determination above 0.999 enable beacon-field geometry selection considering the height limit and fault tolerance.
No takes yet. Share an insight, caveat, or question.
Zhevzhyk et al. (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: