Audiovisual time-to-collision estimation for accelerating vehicles: The acoustic signature of electric vehicles impairs pedestrians' judgments

Wessels, Marlene; Kröling, Sophie; Oberfeld, Daniel · 2022 · Crossref

DOI: 10.1016/j.trf.2022.09.023

archive: archived pipeline: cataloged verified

Get this paper ↗ (DOI — opens at the source; we link to it, we don't host it)

Summary

This study investigates how the acoustic signatures of electric vehicles (EVs) affect pedestrians' ability to estimate time-to-collision (TTC) compared to internal combustion engine vehicles (ICEVs). The research addresses a critical safety gap: while EVs are quieter, their lack of salient auditory cues during acceleration may impair pedestrians' judgments, potentially leading to risky road-crossing decisions. Specifically, the authors examined whether the altered acoustic profile of EVs, with and without Active Acoustic Vehicle Alerting Systems (AVAS), results in TTC estimation errors similar to those observed in visual-only conditions, where humans typically fail to account for acceleration. To test this, the researchers employed a novel interactive audiovisual virtual reality (VR) system. Auditory stimuli were derived from real-world recordings of an ICEV (Kia Rio) and an EV (Kia e-Niro) with AVAS activated or deactivated, captured on a test track using precise GPS tracking. These recordings were processed through acoustic VR software (TASCAR) to simulate dynamic spatial sound fields, including Doppler effects and reflections, presented via a 16-loudspeaker array using higher-order Ambisonics. Visual stimuli were displayed on a head-mounted display, modeling an urban street scene. Participants estimated the TTC of approaching vehicles by pressing a button when they believed the vehicle would reach a specific point after it was occluded. The experiment included 93 conditions varying by vehicle type, velocity (constant or accelerating), and actual TTC. The results demonstrated that for vehicles traveling at constant speeds, TTC estimates were accurate and similar across all vehicle types. However, for accelerating vehicles, vehicle type significantly impacted estimation accuracy. Pedestrians substantially overestimated the TTC of accelerating EVs, perceiving them as arriving later than they actually did. This overestimation increased with higher acceleration levels and longer actual TTCs, mirroring a "first-order" estimation pattern where acceleration is ignored. In contrast, TTC estimates for accelerating ICEVs were largely accurate, as their salient engine sounds provided clear cues about acceleration. The inclusion of AVAS in EVs somewhat improved TTC estimates but did not restore accuracy to the level observed with ICEVs. These findings indicate that the acoustic signature of EVs impairs pedestrians' ability to judge the arrival time of accelerating vehicles, creating a collision risk. The study concludes that current AVAS technologies are insufficient to fully compensate for the lack of natural acceleration cues found in ICEVs. This has significant implications for road traffic safety and suggests that future AVAS designs must more effectively convey acceleration information to ensure pedestrians can make safe crossing decisions.

Provenance

The full processing record for this entry. Every stage of this paper's journey through the pipeline is logged — what ran, with which tool and model, how many attempts it took, and when it last completed.

Summary generated by qwen3.6-27b-prismaquant on 2026-06-25; verification: verified.

Topics

Ranked by relevance to this paper. Hover a topic for its definition.

• time to collision looming
• auditory warnings