Interaction-Based Vehicle Automation Model for Intelligent Vision Systems
Electronics · 27.08.2025 · 14 (17) · 3406
DOI: 10.3390/electronics14173406 ↗Kurz gefasst
Fachartikel in der Zeitschrift Electronics, online erschienen am 27.08.2025. Verfasst von Julian Balcerek und 1 weiteren Autor:innen. Die Arbeit wurde bisher 2-mal zitiert (Crossref) und nennt 53 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Poznan University of Technology.
Abstract (Originalsprache)
In this paper, we introduce a new vehicle automation model describing the latest intelligent vision systems, but not limited to them, based on interactions between the vehicle, its user, and the environment, occurring simultaneously or at different times. The proposed model addresses the lack of vehicle automation models that would simultaneously incorporate the latest vision systems and human actions, organize them according to interaction types, and enable quantitative performance analysis. The model, based on interactions, organizes in terms of types and enables parametric analysis of the operation of the latest automatic vision systems and modern knowledge about human behavior using the perception of visual information. The concept of interaction cycles was introduced, thanks to which it is possible to analyze subsequently occurring interactions, i.e., when actions trigger reactions. Interactions were decomposed into fragments containing single direct unidirectional interactions. The interactions have been assigned consistent numerical effectiveness parameters related to image recognition by individual systems, thanks to which numerical analysis at different levels of detail is possible, depending on the needs. For each of the six interaction types, ten applications of the newest and available vision systems, including those prepared by the authors, were reviewed and selected for effectiveness analysis using the presented model. The analysis was performed by appropriately weighting and averaging or multiplying interaction effectiveness. The overall effectiveness of the interaction model for selected solutions was over 80%. The model also allows the selection of weights for individual components, depending on the criterion being analyzed, e.g., safety or environmental protection. Humans turned out to be the weakest link in interactions, e.g., reducing the human driver role increased overall effectiveness of interactions to almost 90%, and its increase resulted in an effectiveness reduction to over 70%. Examples of selecting solutions for implementation based on the interaction cycle and its fragment, taking into account the effectiveness of subsequent interactions, were also presented. The presented model is characterized by its comprehensiveness, simplicity and scalability according to needs. It can be used both for scientific analysis of existing solutions and be helpful in selecting solutions for modification and implementation by the vehicle manufacturers.
Förderung
- Poznan University of Technology
Zitieren
APA
Balcerek, J., & Pawłowski, P. (2025). Interaction-Based Vehicle Automation Model for Intelligent Vision Systems. Electronics, 14(17), 3406. https://doi.org/10.3390/electronics14173406
Vancouver
Balcerek J, Pawłowski P. Interaction-Based Vehicle Automation Model for Intelligent Vision Systems. Electronics. 2025;14(17):3406. doi:10.3390/electronics14173406
BibTeX
@article{Balcerek2025,
author = {Balcerek, Julian and Pawłowski, Paweł},
title = {Interaction-Based Vehicle Automation Model for Intelligent Vision Systems},
journal = {Electronics},
year = {2025},
volume = {14},
number = {17},
pages = {3406},
doi = {10.3390/electronics14173406}
}
EUVIMED