Applying Conformal Prediction to Uncertainty Quantification of Vision-Based Aircraft Surface Navigation

Abhijit Sarkar, Surendrabikram Thapa, Jin Woo Baik, Mathieu Joerger, and Jason Rife

Peer Reviewed

Abstract: Reliable ground navigation for uncrewed aircraft systems (UAS) operating in airport environments is increasingly challenged by GNSS vulnerability, including signal degradation due to antenna obstruction to the sky, and interference such as spoofing. Vision-based tracking of taxiway and runway markings has emerged as a promising complementary source of relative positioning information, as these features encode structured geometric constraints that can support lateral and heading estimation during ground operations. However, pavement markings are not purely deterministic references; they are perceptually extracted features whose model parameter estimation introduce uncertainty that is rarely modeled within navigation frameworks. This paper investigates conformal prediction as a model-agnostic uncertainty quantification framework for vision-based pavement-marking detection. Rather than proposing a new detector, we evaluate how uncertainty can be calibrated for multiple classes of detection methods, including a classical Canny-Hough pipeline, a learning-based lane-line detector, and a segmentation-based model. For line-based methods, we define a geometric nonconformity score based on the minimum mean absolute distance between the annotated lane and the set of predicted candidates. For segmentation, we use a lane-pixel miscoverage score to construct conformal prediction regions with controlled coverage. Experiments on annotated airport surface imagery show that the proposed framework separates distinct sources of perception uncertainty: candidate-generation failure, geometric mismatch after detection, and conservative prediction sets required for high coverage. The Hough-based method can produce accurate candidates but suffers from limited candidate availability, while the learning-based method produces candidates consistently but requires substantially larger error bounds for high coverage. The segmentation-based approach provides dense uncertainty regions with an explicit coverage-compactness trade-off. These results demonstrate how conformal uncertainty bounds can support integrity-aware navigation by linking vision-based pavement-marking measurements to protection-level and alert-limit reasoning. This study provides a preliminary step toward calibrated, risk-aware use of computer vision for aircraft surface navigation.
Published in: Proceedings of the 39th International Technical Meeting of the Satellite Division of the Institute of Navigation (ION GNSS+ 2026)
September 14 - 18, 2026
Hyatt Regency Grand Cypress
Orlando, Florida
Pages: 798 - 811
Cite this article: Sarkar, Abhijit, Thapa, Surendrabikram, Baik, Jin Woo, Joerger, Mathieu, Rife, Jason, "Applying Conformal Prediction to Uncertainty Quantification of Vision-Based Aircraft Surface Navigation," Proceedings of the 39th International Technical Meeting of the Satellite Division of the Institute of Navigation (ION GNSS+ 2026), Orlando, Florida, September 2026, pp. 798-811. https://doi.org/10.33012/2026.20852
Full Paper:

Sign in to download with your ION membership, download credits or subscription, or buy this article.

Sign In Buy this article: $12 for ION members, $35 for non-members
References:
Show 36 references
  1. Angelopoulos, A. N. and Bates, S. (2023). Conformal prediction: A gentle introduction. Foundations and Trends in Machine Learning, 16(4):494-591.
  2. Angelopoulos, A. N., Bates, S., Fisch, A., Lei, L., and Schuster, T. (2024). Conformal risk control. In International Conference on Learning Representations.
  3. Canny, J. (1986). A computational approach to edge detection. IEEE Transactions on Pattern Analysis and Machine Intelligence, PAMI-8(6):679-698.
  4. DeCleene, B. (2000). Defining pseudorange integrity-overbounding. In Proceedings of the 13th international technical meeting of the satellite division of the institute of navigation (ION GPS 2000), pages 1916-1924.
  5. DoT (2025). The advanced air mobility national strategy. https://www.transportation.gov/aam#departure. Accessed: 2026-06-24.
  6. Duda, R. O. and Hart, P. E. (1972). Use of the hough transformation to detect lines and curves in pictures. Communications of the ACM, 15(1):11-15.
  7. Eaton, W. H. (2017). Automated taxiing for unmanned aircraft systems. PhD thesis, Loughborough University.
  8. Felux, M., Figuet, B., Waltert, M., Fol, P., Strohmeier, M., and Olive, X. (2023). Analysis of gnss disruptions in european airspace. In Proceedings of the 2023 International Technical Meeting of The Institute of Navigation, pages 315-326.
  9. Gal, Y. and Ghahramani, Z. (2016). Dropout as a bayesian approximation: Representing model uncertainty in deep learning. In Proceedings of the International Conference on Machine Learning, pages 1050-1059.
  10. PMLR.
  11. Gruyer, D., Belaroussi, R., and Revilloud, M. (2016). Accurate lateral positioning from map data and road marking detection. Expert Systems with Applications, 43:1-8.
  12. Hou, Y., Ma, Z., Liu, C., and Loy, C. C. (2019). Learning lightweight lane detection cnns by self attention distillation. In Proceedings of the IEEE/CVF International Conference on Computer Vision, pages 1013-1021.
  13. Joerger, M. and Pervan, B. (2019). Quantifying safety of laser-based navigation. IEEE Transactions on Aerospace and Electronic Systems, 55(1):273-288.
  14. Kawamura, E., Kannan, K., Lombaerts, T., and Ippolito, C. A. (2022). Vision-based precision approach and landing for advanced air mobility. In AIAA SciTech 2022 Forum, page 0497.
  15. Kendall, A. and Gal, Y. (2017). What uncertainties do we need in bayesian deep learning for computer vision? In Advances in Neural Information Processing Systems, volume 30.
  16. Kigotho, O. N. and Rife, J. H. (2021). Comparison of rectangular and elliptical alert limits for lane-keeping applications. In Proceedings of the 34th International Technical Meeting of the Satellite Division of The Institute of Navigation (ION GNSS+ 2021), pages 93-104.
  17. Kim, D., Kim, B., Chung, T., and Yi, K. (2016). Lane-level localization using an avm camera for an automated driving vehicle in urban environments. IEEE/ASME Transactions on Mechatronics, 22(1):280-290.
  18. Lakshminarayanan, B., Pritzel, A., and Blundell, C. (2017). Simple and scalable predictive uncertainty estimation using deep ensembles. In Advances in Neural Information Processing Systems, volume 30.
  19. Murrian, M. J., Narula, L., and Humphreys, T. E. (2019). Characterizing terrestrial gnss interference from low earth orbit. In Proceedings of the 32nd International Technical Meeting of the Satellite Division of The Institute of Navigation (ION GNSS+ 2019), pages 3239-3253.
  20. Osechas, O., Fohlmeister, F., Dautermann, T., and Felux, M. (2022). Impact of gnss-band radio interference on operational avionics. NAVIGATION: Journal of the Institute of Navigation, 69(2).
  21. Pan, X., Shi, J., Luo, P., Wang, X., and Tang, X. (2018). Spatial as deep: Spatial cnn for traffic scene understanding. In Proceedings of the AAAI conference on artificial intelligence, volume 32.
  22. Qin, Z., Wang, H., and Li, X. (2020a). Ultra fast structure-aware deep lane detection. In Computer Vision -ECCV 2020, pages 276-291. Springer.
  23. Qin, Z., Wang, H., and Li, X. (2020b). Ultra fast structure-aware deep lane detection. In Proceedings of the European Conference on Computer Vision, pages 276-291. Springer.
  24. Ravi, N., Gabeur, V., Hu, Y.-T., Hu, R., Ryali, C., Ma, T., Khedr, H., Rädle, R., Rolland, C., Gustafson, L., et al. (2025). Sam 2: Segment anything in images and videos. In International Conference on Learning Representations, volume 2025, pages 28085-28128.
  25. Rife, J. and Pullen, S. (2009). Aviation Applications, chapter 10, pages 252-257.
  26. Rife, J., Pullen, S., Enge, P., and Pervan, B. (2006). Paired overbounding for nonideal laas and waas error distributions. IEEE Transactions on Aerospace and Electronic Systems, 42(4):1386-1395.
  27. Romano, Y., Patterson, E., and Candès, E. J. (2019). Conformalized quantile regression. In Advances in Neural Information Processing Systems, volume 32, pages 3543-3553.
  28. Rose, C., Britt, J., Allen, J., and Bevly, D. (2014). An integrated vehicle navigation system utilizing and lateral position estimation systems in difficult environments for gps. IEEE Transactions on Intelligent Transportation Systems, 15(6):2615-2629.
  29. RTCA (2021). SC-228: Minimum operational performance standards (MOPS) for detect and avoid (DAA) systems. Technical Report DO-365B, RTCA, Inc., Washington, DC.
  30. Simon, D. (2006). Optimal State Estimation: Kalman, H ? , and Nonlinear Approaches. Wiley-Interscience, Hoboken, NJ.
  31. Tabelini, L., Berriel, R., Paixão, T. M., Badue, C., De Souza, A. F., and Oliveira-Santos, T. (2021). Keep your eyes on the lane: Real-time attention-guided lane detection. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pages 294-302.
  32. Vovk, V., Gammerman, A., and Shafer, G. (2005). Algorithmic Learning in a Random World. Springer, New York.
  33. Walter, T. and Enge, P. (2004). Modernizing waas. In Proceedings of the 17th International Technical Meeting of the Satellite Division of the Institute of Navigation (ION GNSS 2004), pages 1683-1690.
  34. Zhang, S., Liu, H., Wang, H., Fang, K., and Zhong, K. (2025). Aircraft taxi guidance and positioning method based on onboard forward view cameras. In Proceedings of the 2025 International Technical Meeting of The Institute of Navigation, pages 992-1007.
  35. Zheng, T., Fang, H., Zhang, Y., Tang, W., Yang, Z., Liu, H., and Cai, D. (2021). Resa: Recurrent feature-shift aggregator for lane detection. In Proceedings of the AAAI conference on artificial intelligence, volume 35, pages 3547-3554.
  36. Zheng, T., Huang, Y., Liu, Y., Tang, W., Yang, Z., Cai, D., and He, X. (2022). Clrnet: Cross layer refinement network for lane detection. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pages 898-907.