- Titel
- Interpretable Passenger Discomfort Prediction in Automated Driving Using Transformers
- Zitierfähige Url:
- https://nbn-resolving.org/urn:nbn:de:bsz:ch1-qucosa2-1005735
- Konferenz
- 9th Humanist Virtual Centre of Excellence Conference 2025. Chemnitz, 27.-29. August 2025
- Quellenangabe
- The 9th HUMANIST Conference - Virtual Centre of Excellence - Chemnitz, Germany 27–29 August 2025
- DOI
- https://doi.org/10.60687/2025-0214
- Abstract (EN)
- This study employs a Temporal Fusion Transformer (TFT) model to forecast passenger discomfort in the context of highly automated driving, utilizing time-series data collected from a driving simulator. The objective is to assess whether transformer-based models can provide accurate and interpretable predictions of subjective discomfort levels during vehicle driving. Two variants of the TFT model were developed, referred to as TFT-full and TFT-restricted. Both models were trained on simulator data from 100 participants. Additionally, they were compared to DeepAR, an autoregressive time-series model used as a baseline. Results showed that the TFT-restricted model performed better than both the TFT-full and DeepAR models. The TFT-restricted model achieved the best overall performance, with a Mean Absolute Error (MAE) of 0.042 and a Root Mean Square Error (RMSE) of 0.133. These findings support the suitability of transformer-based models for discomfort detection in automated vehicles.
- Freie Schlagwörter (EN)
- Discomfort detection, human technology interaction, Transformer
- Klassifikation (DDC)
- 158
- Normschlagwörter (GND)
- Mensch-Maschine-Interaktion, Transformer
- Herausgeber (Institution)
- Technische Universität Chemnitz, Chemnitz
- Version / Begutachtungsstatus
- publizierte Version / Verlagsversion
- URN Qucosa
- urn:nbn:de:bsz:ch1-qucosa2-1005735
- Veröffentlichungsdatum Qucosa
- 01.12.2025
- Dokumenttyp
- Konferenzbeitrag
- Sprache des Dokumentes
- Englisch
- Lizenz / Rechtehinweis
CC BY-SA 4.0