Assessment of external interface of autonomous vehicles

Orlický, Adam; Mashko, Alina; Mík, Josef · 2021 · DOAJ

DOI: 10.14311/AP.2021.61.0733

archive: archived pipeline: cataloged verified

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

Summary

This paper addresses the critical safety challenge of communication between autonomous vehicles (AVs) and pedestrians, particularly at unregulated crossings. As AVs eliminate the human driver, traditional non-verbal cues like eye contact or hand gestures are lost, potentially leading to misunderstandings and accidents. The authors highlight that pedestrians are vulnerable road users, with survival probabilities dropping significantly at higher impact speeds. To mitigate this, the study aims to develop and validate a methodology for assessing external Human-Machine Interfaces (e-HMI)—visual signals displayed on the vehicle—to ensure they effectively convey intent and improve pedestrian trust and safety. The researchers employed a virtual reality (VR) experimental design to test pedestrian behavior, chosen for its repeatability, safety, and ability to control variables compared to naturalistic studies. The experiment utilized an HTC Vive Pro Eye headset with integrated eye-tracking technology to monitor subjects' gaze behavior. The simulated scenario involved a two-lane urban road with a zebra crossing. Subjects were tasked with crossing the road when they felt it was safe. The study tested four conditions by varying braking behavior (normal vs. aggressive) and the presence of an e-HMI (specifically, LED signals on the windshield, based on a Ford concept). "Normal" braking began at 50 meters, while "aggressive" braking began at 35 meters. Data collected included the distance of the vehicle when the subject first looked at it, the distance when the subject began crossing, and the duration of the first glance. The results, derived from 18 subjects, demonstrated that the presence of an e-HMI significantly improved pedestrian decision-making. Subjects crossed the road sooner and at a safer distance from the vehicle when visual signals were present. Specifically, the e-HMI reduced the time required to evaluate the situation, evidenced by shorter first-glance durations. In aggressive braking scenarios, the waiting time decreased by a factor of eight when the interface was present. Furthermore, subjects recognized the visual message at a closer average distance (9.6 m) compared to scenarios without signals (15 m). While subjective reports indicated that many participants did not consciously distinguish between braking modes, objective data showed that 56% entered the crossing sooner during normal deceleration, indicating that the e-HMI effectively compensated for the lack of driver cues and reduced mental load. The study concludes that external communication interfaces are essential for replacing human-driver interactions in autonomous driving, thereby enhancing traffic safety for vulnerable road users. The proposed VR-based methodology provides a valid, efficient tool for assessing and designing e-HMI systems, allowing for the testing of various interface designs and placements. The findings suggest that standardized visual signals can increase pedestrian trust and reduce decision times, contributing to the broader standardization of AV communication. Future research should expand the sample size and integrate qualitative studies to further refine the objective assessment tools.

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.

StageOutcomeToolModelPromptAttemptsCompleted
discover success DOAJ 1 2026-06-25
archive success unpaywall 1 2026-06-26
extract success cached 2 2026-06-26
clean success clean 1 2026-06-25
chunk success chunk 1 2026-06-25
embed success embed Qwen/Qwen3-Embedding-8B 1 2026-06-25
promote success 1 2026-06-25
summarize success llm qwen3.6-27b-prismaquant summ-v5 1 2026-06-26
tag success vector_similarity 6 2026-06-25
verify partial 1 2026-06-26

Summary generated by qwen3.6-27b-prismaquant on 2026-06-26; verification: verified_with_issues.

Topics

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

Information type

What kind of knowledge this paper contributes, grouped by family — independent of topic (what it is about) and method (how it was studied).