Validation of an urban environment for pedestrian behavior analysis in full immersive virtual reality
DOI: 10.1016/j.trpro.2024.02.004
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 addresses the critical need for safe and effective methods to analyze pedestrian behavior, particularly among children, who are vulnerable road users with high mortality rates from traffic crashes. While virtual reality (VR) has emerged as a valuable tool for creating controlled, non-existent, or dangerous scenarios, its validity for studying naturalistic pedestrian behavior remains underexplored. The primary objective of this research was to validate a full-immersive VR environment as a reliable tool for investigating pedestrian kinematics and decision-making. The authors aimed to determine if VR-generated data aligns with real-world naturalistic studies and to identify which environmental factors influence children’s walking speeds. The experimental design involved 122 middle school students (aged 11–14) from a school in Italy. Participants wore HP Reverb VR headsets connected to a backpack PC, allowing them to move freely within a 22m x 12m gymnasium. The study utilized Unity® software to create 15 distinct urban scenarios: nine road-crossing trials and six strolling trials. Crossing scenarios varied by intersection type (signalized vs. non-signalized), traffic presence, and light conditions. Strolling scenarios included straight paths with potential hazards like opening car doors or emerging e-scooters. Kinematic data, including speed, position, and head rotation, were collected at 25 Hz. Statistical analyses, including one-way ANOVA, one-sample t-tests, and linear mixed models, were performed to compare VR data against established literature values and to assess the impact of personal characteristics and scenario variables on behavior. The results demonstrated that the VR environment successfully replicated real-world pedestrian behavior. The average crossing speed in VR (1.358 m/s) showed no significant difference from speeds reported in naturalistic studies (1.34–1.36 m/s). Participants walked significantly faster during crossing scenarios than during strolling scenarios (1.040 m/s), consistent with real-world observations where pedestrians accelerate to minimize time in the roadway. Crucially, personal characteristics such as age, gender, and mobility habits had no significant correlation with walking speed. Instead, environmental factors drove behavioral changes. Higher speeds were observed in non-signalized crossings and scenarios with moving traffic, while the lowest speeds occurred at signalized crossings with red lights. This indicates that children perceived signalized intersections as safer, leading to more cautious behavior, whereas non-signalized contexts induced riskier, faster movements. The significance of this work lies in the validation of immersive VR as a robust tool for pedestrian safety research and education. By confirming that VR data aligns with naturalistic benchmarks, the study supports the use of VR for investigating dangerous scenarios that are difficult or unethical to replicate in the real world. The findings highlight that pedestrian behavior is heavily influenced by infrastructure design rather than individual traits, underscoring the importance of designing intersections that mitigate risk. Furthermore, the validated tool offers a platform for future research into road safety education, hazard perception, and the evaluation of infrastructure interventions aimed at reducing crash risks for vulnerable users.
Key finding
The average walking speeds of children in the virtual reality environment were statistically consistent with real-world naturalistic studies, validating the tool's realism and revealing that non-signalized crossings elicited higher speeds than signalized ones.
Methodology
simulator
Sample size: 122
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. Discovered via author_sweep_intake on 2026-05-28.
| Stage | Outcome | Tool | Model | Prompt | Attempts | Completed |
|---|---|---|---|---|---|---|
| discover | success | author_sweep | — | — | 2 | 2026-05-28 |
| archive | success | canonical_url | — | — | 11 | 2026-06-06 |
| extract | success | cached | — | — | 3 | 2026-06-10 |
| clean | success | clean | — | — | 1 | 2026-06-04 |
| chunk | success | chunk | — | — | 1 | 2026-06-04 |
| embed | success | embed | Qwen/Qwen3-Embedding-8B | — | 1 | 2026-06-04 |
| enrich | success | — | — | — | 1 | 2026-05-28 |
| promote | success | — | — | — | 1 | 2026-06-04 |
| summarize | success | llm | qwen3.6-27b-prismaquant | summ-v5 | 2 | 2026-06-10 |
| tag | success | vector_similarity | — | — | 15 | 2026-06-11 |
| verify | success | — | — | — | 2 | 2026-06-10 |
Summary generated by qwen3.6-27b-prismaquant on 2026-06-10; verification: verified.
Topics
Ranked by relevance to this paper. Hover a topic for its definition.