Online Map Validation for Autonomous Driving

ACM Computer Science in Cars Symposium (CSCS 2019)

autonomous-driving
map-validation
Two complementary approaches for online HD map validation: a model-based framework and a deep similarity learning technique, enabling real-time detection of map inconsistencies from live sensor data.
Authors

Andrea Fabris

Felix Drost

Luca Parolini

Qing Rao

Andreas Rauch

Sebastian Schneider

Sebastian Wagner

Alois Knoll

Published year

2019

Authors
Andrea Fabris, Felix Drost, Luca Parolini, Qing Rao, Andreas Rauch, Sebastian Schneider, Sebastian Wagner, Alois Knoll
Published
CSCS
Preprint
PDF
Poster
PDF

Overview of the map validation pipeline

Abstract

Data in HD maps used by autonomous vehicles can be outdated and erroneous. This paper proposes two complementary approaches for online map validation. One uses model-based frameworks while the other applies deep similarity learning techniques. Together they allow the vehicle to detect map inconsistencies in real time from live sensor data, enabling safer autonomous operation.

Related

Citation

BibTeX citation:
@inproceedings{fabris2019,
  author = {Fabris, Andrea and Drost, Felix and Parolini, Luca and Rao,
    Qing and Rauch, Andreas and Schneider, Sebastian and Wagner,
    Sebastian and Knoll, Alois},
  title = {Online {Map} {Validation} for {Autonomous} {Driving}},
  booktitle = {ACM Computer Science in Cars Symposium (CSCS)},
  date = {2019-10-01},
  url = {https://lucaparolini.com/publications/papers/online-map-validation-cscs-2019/},
  langid = {en}
}
For attribution, please cite this work as:
A. Fabris et al., “Online Map Validation for Autonomous Driving,” in ACM Computer Science in Cars Symposium (CSCS), Oct. 2019. Available: https://lucaparolini.com/publications/papers/online-map-validation-cscs-2019/