FIM-optimized staggered sensor geometry combined with Phy-GAANet using physics-informed features and geometry-aware attention achieves 1.84 mm position error and 3.18 degree orientation error at over 270 Hz in real-world tests, outperforming classical solvers and standard CNNs.
Pr- posnet: A cnn-transformer model with coordinate attention for precision and robust magnetic pose estimation
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.RO 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Information-Theoretic Geometry Optimization and Physics-Aware Learning for Calibration-Free Magnetic Localization
FIM-optimized staggered sensor geometry combined with Phy-GAANet using physics-informed features and geometry-aware attention achieves 1.84 mm position error and 3.18 degree orientation error at over 270 Hz in real-world tests, outperforming classical solvers and standard CNNs.