SafetyQuantifiable-PLVINS

Xi Zheng November 11, 2025

Safety-quantifiable Line Feature-based Monocular Visual Localization with 3D Prior Map

To address the drift and safety quantification challenges in visual localization, we propose a novel map-aided method that delivers both accurate pose estimates and a measurable error bound. By tightly integrating visual-inertial odometry with a prior line map, our system establishes geometric constraints between 2D image features and 3D map lines. Crucially, we introduce a GNSS-inspired integrity monitoring framework to compute a Protection Level (PL), which quantifies the potential error in both position and orientation, thereby certifying the solution’s safety.

For details, please refer to our official repository at SafetyQuantifiable-PLVINS

And if you are using this code, please cite our paper by

@article{zheng2025safety,
  title={Safety-quantifiable line feature-based monocular visual localization with 3d prior map},
  author={Zheng, Xi and Wen, Weisong and Hsu, Li-Ta},
  journal={IEEE Transactions on Intelligent Transportation Systems},
  year={2025},
  publisher={IEEE}
}