Skip to main content

Research Repository

Advanced Search

Explainable machine learning for autonomous vehicle positioning using SHAP

Onyekpe, Uche; Lu, Yang; Apostolopoulou, Eleni; Palade, Vasile; Eyo, Eyo Umo; Kanarachos, Stratis

Explainable machine learning for autonomous vehicle positioning using SHAP Thumbnail


Authors

Uche Onyekpe

Yang Lu

Eleni Apostolopoulou

Vasile Palade

Profile image of Eyo Eyo

Dr Eyo Eyo Eyo.Eyo@uwe.ac.uk
Lecturer in Geotechnical Engineering

Stratis Kanarachos



Contributors

Mayuri Mehta
Editor

Vasile Palade
Editor

Indranath Chatterjee
Editor

Abstract

Despite the recent advancements in Autonomous Vehicle (AV) technology, safety still remains a key challenge for their commercialisation and development. One of the major systems influencing the safety of AVs is its navigation system. Road localisation of autonomous vehicles is reliant on consistent accurate Global Navigation Satellite System (GNSS) positioning information. The GNSS relies on a number of satellites to perform triangulation and may experience signal loss around tall buildings, bridges, tunnels, trees, etc. We previously proposed the Wheel Odometry Neural Network (WhONet) as an approach to provide continuous positioning information in the absence of the GNSS signals. We achieved this by integrating the GNSS output with the wheel encoders’ measurements from the vehicle whilst also learning the uncertainties present in the position estimation. However, the positioning problem is a safety critical one and thus requires a qualitative assessment of the reasons for the predictions of the WhONet model at any point of use. There is therefore the need to provide explanations for the WhONet’s predictions to justify its reliability and thus provide a higher level of transparency and accountability to relevant stakeholders. Explainability in this work is achieved through the use of Shapley Additive exPlanations (SHAP) to examine the decision-making process of the WhONet model on an Inertial and Odometry Vehicle Navigation Benchmark Data subset describing an approximate straight-line trajectory. Our study shows that on an approximate straight-line motion, the two rear wheels are responsible for the most increase in the position uncertainty estimation error compared to the two front wheels.

Online Publication Date Oct 20, 2022
Publication Date Jan 1, 2023
Deposit Date Oct 28, 2022
Publicly Available Date Oct 21, 2024
Publisher Springer
Pages 157-183
Series Title Part of the Intelligent Systems Reference Library book series (ISRL,volume 232)
Series ISSN 1868-4408; 1868-4394
Edition 1st
Book Title Explainable AI: Foundations, Methodologies and Applications
Chapter Number 8
ISBN 9783031128066; 9783031128073
DOI https://doi.org/10.1007/978-3-031-12807-3_8
Keywords Wheel odometry, Autonomous vehicles, Inertial navigation system, Deep learning, Explainable machine learning, GNSS outage, Positioning, Neural networks
Public URL https://uwe-repository.worktribe.com/output/10106907
Publisher URL https://link.springer.com/chapter/10.1007/978-3-031-12807-3_8
Related Public URLs https://link.springer.com/book/10.1007/978-3-031-12807-3

https://www.springer.com/series/8578
Additional Information First Online: 20 October 2022

Files

Explainable machine learning for autonomous vehicle positioning using SHAP (1.1 Mb)
PDF

Licence
http://www.rioxx.net/licenses/all-rights-reserved

Publisher Licence URL
http://www.rioxx.net/licenses/all-rights-reserved

Copyright Statement
This is the author’s accepted manuscript of their book chapter 'Explainable machine learning for autonomous vehicle positioning using SHAP', which features in the book 'Explainable AI: Foundations, Methodologies and Applications' published by Springer (Copyright Springer 2023). The final published version is available here: https://link.springer.com/chapter/10.1007/978-3-031-12807-3_8





You might also like



Downloadable Citations