Perceived risk and acceptance of automated vehicles users to unexpected hazard situations in real driving conditions
The decision-making systems in highly automated vehicles use artificial intelligence (AI) techniques to replicate human-like behaviour in driving. However, when studying the acceptance of such vehicles, analysing the human factor is essential if their deployment is to be a success. This paper presen...
| Authors: | , , , , , , , |
|---|---|
| Format: | article |
| Publication Date: | 2025 |
| Country: | España |
| Institution: | Universidad Complutense de Madrid (UCM) |
| Repository: | Docta Complutense |
| Language: | English |
| OAI Identifier: | oai:docta.ucm.es:20.500.14352/129918 |
| Online Access: | https://hdl.handle.net/20.500.14352/129918 |
| Access Level: | Open access |
| Keyword: | Automated vehicles Perceived risk Users acceptance Pupil dilation Safety Unexpected hazards Psicología experimental 6106 Psicología Experimental |
| Summary: | The decision-making systems in highly automated vehicles use artificial intelligence (AI) techniques to replicate human-like behaviour in driving. However, when studying the acceptance of such vehicles, analysing the human factor is essential if their deployment is to be a success. This paper presents an analysis, based on real driving conditions, of perceived risk and occupant acceptance of an SAE level 4 automated vehicle (where the user adopts only the role of a passenger) in two unforeseen situations involving risk, one in which the life of a pedestrian may be endangered and the other in which the life of the user himself may be endangered. To assess perceived risk, pupil diameter, a physiological variable closely related to emotional arousal, was measured. The study involved 40 participants. The study revealed that perceived risk is significantly higher when the user's personal safety is compromised, reinforcing the principle of prioritising passenger safety in autonomous vehicle design. Key findings include an inverse correlation between perceived risk and driving experience, and no evidence suggesting that prior automated vehicle experience modifies future usage expectations. The research emphasises the critical role of human-centered design in AI-driven decision-making systems, highlighting the importance of understanding psychological and physiological responses to autonomous technology. |
|---|