Perceived risk evolution in automated driving inferred from large-scale discrete ratings

Xiaolin He, Zirui Li, Xinwei Wang, Riender Happee, Meng Wang · 2025

Read Perceived risk evolution in automated driving inferred from large-scale discrete ratings by Xiaolin He, Zirui Li, Xinwei Wang, Riender Happee, Meng Wang (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2508-19121.txt.

Introduction

He and co-authors reconstruct how fear rises and falls inside a self-driving car. Passengers rate their perceived risk clip by clip, but single scores hide the difference between a brief spike and a long climb. Treating the discrete ratings as constraints on a hidden curve, the team infers the full time course of perceived risk. The data span more than two thousand riders and 236 hours of scripted motorway driving.

The inferred curves separate total accumulated risk from how concentrated it is within a clip, distinctions peak scores cannot make. A neural network then predicts the curves from vehicle and traffic cues. Attribution analysis shows conflict cues dominate while risk rises, and stability cues take over as it falls. The authors offer the method as a finer lens on rider experience.

Catalog record

CREATORXiaolin He, Zirui Li, Xinwei Wang, Riender Happee, Meng Wang
YEAR2025
COLLECTIONPAPERS
FORMATTXT
FILE SIZE154 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2508-19121

Browse the full library · About the Tower of Babel