Training Drivers to Automated Vehicles

Mangharam, Rahul; Loeb, Helen; Qiao, Zhijie · 2023 · ROSA P / Mobility21, Carnegie Mellon University

archive: archived pipeline: cataloged verified

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

Summary

This research report addresses the critical barrier to widespread autonomous vehicle (AV) adoption: public mistrust and apprehension. Despite the proven safety benefits of AVs, surveys from AAA, Pew Research, and PAVE indicate that a majority of Americans distrust the technology, largely due to a lack of understanding regarding AV capabilities and ethical decision-making. The authors hypothesize that immersive training can bridge this gap by helping drivers understand AV operational limits and building comfort with relinquishing control. The study aims to test the efficacy of a virtual reality (VR) driving simulator as a cost-effective, safe, and scalable tool for AV education and demonstration. The researchers developed a VR-based simulation using the open-source CARLA driving simulator, modified to mimic Level 4 autonomous driving capabilities. The system was coupled with a Logitech G29 steering wheel and pedal set to enhance fidelity. Participants, defined as drivers with conventional vehicle experience but limited AV knowledge, underwent a pre- and post-simulation survey protocol. The simulation exposed users to rural, city, and highway scenarios, allowing them to experience the AV’s driving style and practice takeover procedures. The study sought to answer quantitative questions regarding whether the simulator decreased perceived risk and increased perceived usefulness, ease-of-use, trust, and behavioral intentions toward AVs, as well as qualitative questions about the suitability of simulators for dealership demonstrations and driving school curricula. The findings demonstrate that simulator-based education effectively alters driver perceptions. The study confirms that immersive, hands-on experience helps participants form accurate mental models of AV behavior, thereby reducing the bias that often leads individuals to undervalue AV performance compared to human drivers. By providing a first-person perspective on AV operation, the simulator addresses the "human factor" of resistance, specifically the fear of losing control. The results align with previous literature suggesting that direct experience, even simulated, is more effective than passive information delivery in building trust. The data supports the conclusion that education significantly correlates with willingness to use AVs, with knowledgeable individuals showing markedly higher acceptance rates than those with no prior exposure. The significance of this work lies in its contribution to AV deployment strategies and policy. The authors argue that driving simulators offer a superior alternative to expensive and logistically complex real-world test drives for public education. The study provides a systematic approach for integrating CARLA into VR frameworks, making it accessible for broader application. The implications suggest that policymakers, auto dealerships, and driving schools should adopt simulator-based training to accelerate AV acceptance. By mandating or encouraging such training, society can move closer to "Vision Zero" goals, leveraging the safety benefits of AVs while overcoming the psychological and educational hurdles that currently limit their widespread use.

Key finding

Using a VR driving simulator significantly decreased drivers' perceived risk and increased their trust and perceived usefulness of automated vehicles.

Methodology

simulator

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 bulk_ingest_rosap on 2026-05-23 (6 acquisition events logged).

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.

• acceptance adoption
• automation
• trust calibration
• automation surprise
• situational awareness
• simulator training transfer

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).

• Methodological Resource: tool software, validation psychometrics
• Theoretical Contribution: computational model