Högskolan i Skövde

his.sePublications
Change search
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • apa-cv
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • Other style
More styles
Language
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Other locale
More languages
Output format
  • html
  • text
  • asciidoc
  • rtf
Surrogate Deep Learning to Estimate Uncertainties for Driver Intention Recognition
University of Skövde, School of Informatics. University of Skövde, Informatics Research Environment. Volvo Car Corporation Sweden. (Skövde Artificial Intelligence Lab (SAIL))ORCID iD: 0000-0003-2135-6615
University of Skövde, School of Informatics. University of Skövde, Informatics Research Environment. (Skövde Artificial Intelligence Lab (SAIL))ORCID iD: 0000-0003-2973-3112
University of Skövde, School of Informatics. University of Skövde, Informatics Research Environment. (Skövde Artificial Intelligence Lab (SAIL))ORCID iD: 0000-0003-2949-4123
University of Skövde, School of Informatics. University of Skövde, Informatics Research Environment. (Skövde Artificial Intelligence Lab (SAIL))ORCID iD: 0000-0001-8884-2154
Show others and affiliations
2023 (English)In: ICMLC 2023: Proceedings of 2023 15th International Conference on Machine Learning and Computing, Zhuhai, China, February 17-20, 2023, New York, NY, USA: Association for Computing Machinery (ACM), 2023, p. 252-258Conference paper, Published paper (Refereed)
Abstract [en]

Real-world applications of artificial intelligence that can potentially harm human beings should be able to express uncertainty about the made predictions. Probabilistic deep learning (DL) methods (e.g., variational inference [VI], VI last layer [VI-LL], Monte-Carlo [MC] dropout, stochastic weight averaging - Gaussian [SWA-G], and deep ensembles) can produce a predictive uncertainty but require expensive MC sampling techniques. Therefore, we evaluated if the probabilistic DL methods are uncertain when making incorrect predictions for an open-source driver intention recognition dataset and if a surrogate DL model can reproduce the uncertainty estimates. We found that all probabilistic DL methods are significantly more uncertain when making incorrect predictions at test time, but there are still instances where the models are very certain but completely incorrect. The surrogate DL models trained on the MC dropout and VI uncertainty estimates were capable of reproducing a significantly higher uncertainty estimate when making incorrect predictions.

Place, publisher, year, edition, pages
New York, NY, USA: Association for Computing Machinery (ACM), 2023. p. 252-258
Keywords [en]
Driver intention recognition, probabilistic deep learning, surrogate modeling, uncertainty quantification
National Category
Information Systems Other Computer and Information Science
Research subject
Skövde Artificial Intelligence Lab (SAIL)
Identifiers
URN: urn:nbn:se:his:diva-22851DOI: 10.1145/3587716.3587758Scopus ID: 2-s2.0-85173817744ISBN: 978-1-4503-9841-1 (print)OAI: oai:DiVA.org:his-22851DiVA, id: diva2:1775537
Conference
15th International Conference on Machine Learning and Computing, Zhuhai, China, February 17-20, 2023
Projects
Intention recognition for real-time automotive 3D situation awareness
Funder
Vinnova, 2018-05012
Note

CC BY-NC-SA 4.0

CORRESPONDING AUTHOR: K. VELLENGA (e-mail: koen.vellenga@volvocars.com)

This work was supported by the Intention Recognition in Real Time for Automotive 3D Situation Awareness (IRRA) Project (https://www.vinnova.se/p/intention-recognition-i-realtid-for-automotive-3d-situation-awareness-irra/).

Available from: 2023-06-27 Created: 2023-06-27 Last updated: 2024-04-17Bibliographically approved
In thesis
1. Advancing Deep Learning-based Driver Intention Recognition: Towards a safe integration framework of high-risk AI systems
Open this publication in new window or tab >>Advancing Deep Learning-based Driver Intention Recognition: Towards a safe integration framework of high-risk AI systems
2024 (English)Licentiate thesis, comprehensive summary (Other academic)
Abstract [en]

Progress in artificial intelligence (AI), onboard computation capabilities, and the integration of advanced sensors in cars have facilitated the development of Advanced Driver Assistance Systems (ADAS). These systems aim to continuously minimize human driving errors. {An example application of an ADAS could be to support a human driver by informing if an intended driving maneuver is safe to pursue given the current state of the driving environment. One of the components enabling such an ADAS is recognizing the driver's intentions. Driver intention recognition (DIR) concerns the identification of what driving maneuver a driver aspires to perform in the near future, commonly spanning a few seconds. A challenging aspect of integrating such a system into a car is the ability of the ADAS to handle unseen scenarios. Deploying any AI-based system in an environment where mistakes can cause harm to human beings is considered a high-risk AI system. Upcoming AI regulations require a car manufacturer to motivate the design, performance-complexity trade-off, and the understanding of potential blind spots of a high-risk AI system.} Therefore, this licentiate thesis focuses on AI-based DIR systems and presents an overview of the current state of the DIR research field. Additionally, experimental results are included that demonstrate the process of empirically motivating and evaluating the design of deep neural networks for DIR. To avoid the reliance on sequential Monte Carlo sampling techniques to produce an uncertainty estimation, we evaluated a surrogate model to reproduce uncertainty estimations learned from probabilistic deep-learning models. Lastly, to contextualize the results within the broader scope of safely integrating future high-risk AI-based systems into a car, we propose a foundational conceptual framework.

Place, publisher, year, edition, pages
Skövde: University of Skövde, 2024. p. ix, 116
Series
Dissertation Series ; 60
Keywords
deep learning, uncertainty estimation, driver intention recognition
National Category
Engineering and Technology Information Systems
Research subject
INF302 Autonomous Intelligent Systems; Skövde Artificial Intelligence Lab (SAIL)
Identifiers
urn:nbn:se:his:diva-23726 (URN)978-91-987907-4-0 (ISBN)
Presentation
2024-05-28, D107, Högskolevägen 1, 541 28, Skövde, 13:15 (English)
Opponent
Supervisors
Funder
Vinnova, 2018-05012
Available from: 2024-04-19 Created: 2024-04-17 Last updated: 2024-09-25Bibliographically approved

Open Access in DiVA

fulltext(630 kB)112 downloads
File information
File name FULLTEXT03.pdfFile size 630 kBChecksum SHA-512
301f78c2d7bd09c683adca822ab5a2935be9ff73459ef2b63a4f2f12cfb3f165dd8849f2898a1915319ae4adf5fe0697e888560912758b29529efd7c578ab950
Type fulltextMimetype application/pdf

Other links

Publisher's full textScopushttp://www.icmlc.org/2023.html

Authority records

Vellenga, KoenKarlsson, AlexanderSteinhauer, H. JoeFalkman, Göran

Search in DiVA

By author/editor
Vellenga, KoenKarlsson, AlexanderSteinhauer, H. JoeFalkman, GöranSjögren, Anders
By organisation
School of InformaticsInformatics Research Environment
Information SystemsOther Computer and Information Science

Search outside of DiVA

GoogleGoogle Scholar
Total: 112 downloads
The number of downloads is the sum of all downloads of full texts. It may include eg previous versions that are now no longer available

doi
isbn
urn-nbn

Altmetric score

doi
isbn
urn-nbn
Total: 500 hits
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • apa-cv
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • Other style
More styles
Language
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Other locale
More languages
Output format
  • html
  • text
  • asciidoc
  • rtf