Baidu Maps Tech Team

Oct 23, 2025 · Artificial Intelligence

How LightGBM Boosts Urban GNSS Accuracy by Detecting NLOS Errors

This article presents a reliable NLOS error identification method for GNSS in urban environments, combining fisheye camera and inertial navigation for objective labeling, extracting six signal features, and employing an optimized LightGBM classifier that achieves high precision and real‑time performance, markedly improving positioning accuracy.