Drive Right: Shaping Public’s Trust, Understanding, and Preference Towards Autonomous Vehicles Using a Virtual Reality Driving Simulator

Qiao, Zhijie; Sun, Xiatao; Loeb, Helen; Mangharam, Rahul · 2023 · Crossref

DOI: 10.1109/iv55152.2023.10186710

archive: archived pipeline: cataloged

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

Summary

This paper addresses the challenge of public mistrust and misunderstanding of autonomous vehicles (AVs), which hinders adoption despite the technology’s potential to reduce accidents and improve mobility. The authors propose a virtual reality (VR) driving simulator as a low-cost, safe, and immersive platform for AV education and demonstration. The study aims to determine if such a simulator can systematically improve drivers’ attitudes toward AVs and increase their intention to use them. The methodology involved developing a VR extension of the open-source CARLA simulator, integrating an Oculus Quest 2 headset and a Logitech G29 steering wheel to create a high-fidelity first-person driving experience. The system supported two modes: manual control and Level 4 autonomous driving, where the AV navigates to a destination while handling traffic, yielding, and emergency maneuvers. Thirty-six licensed drivers (mean age 25.5) participated in a pilot study, completing three scenarios: a rural familiarization phase, a complex city navigation phase, and a highway phase with varying AV driving behaviors (cautious, normal, aggressive). Participants completed pre- and post-simulator surveys based on the Technology Acceptance Model, measuring Trust (TR), Perceived Risk (PR), Perceived Usefulness (PU), Perceived Ease-of-Use (PE), and Behavioral Intention (BI) on a five-point Likert scale. Results indicated that the VR simulator successfully increased participants’ understanding and positively shifted their attitudes. Post-simulator scores improved across all categories: Trust rose from a mean of 3.12 to 3.60, Perceived Risk decreased from 3.44 to 2.95, Perceived Usefulness increased from 3.53 to 3.90, Perceived Ease-of-Use rose from 3.35 to 3.72, and Behavioral Intention increased from 3.63 to 3.98. Notably, internal consistency for Perceived Usefulness improved significantly (Cronbach’s α from 0.03 to 0.67), suggesting that the simulation helped participants form a more coherent view of AV utility. The study found that immersive, interactive simulation was more effective than passive demonstrations for building appropriate trust and reducing anxiety. The significance of this work lies in providing a rigorous, reproducible framework for deploying CARLA in VR for human-factor studies. It demonstrates that VR simulators can serve as effective educational tools for driving schools, auto manufacturers, and policymakers to foster rational trust in AVs. By allowing users to experience AV capabilities in a controlled environment, the platform helps bridge the gap between technological capability and public acceptance, addressing the need for "appropriate trust" rather than blind faith or undue fear.

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 qwen3.8-27b-gittensor on 2026-08-23; verification: pending re-verification.

Topics

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

• acceptance adoption

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).

• Empirical Findings: self report data
• Methodological Resource: tool software, validation psychometrics