JOAN: a framework for human-automated vehicle interaction experiments in a virtual reality driving simulator
DOI: 10.21105/joss.04250
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 introduces JOAN, an open-source software framework designed to facilitate human-automated vehicle (AV) interaction experiments within a virtual reality (VR) driving simulator. The development of JOAN addresses the critical need to study human-AV interactions in mixed traffic environments, where manual drivers, autonomous vehicles, and vulnerable road users coexist. While driving simulators are essential for this research, existing state-of-the-art solutions often require expensive, dedicated hardware and possess steep programming learning curves that hinder rapid algorithm development and experiment customization. Conversely, while VR-based simulators offer lower costs and greater flexibility, many lack the necessary frameworks to implement AV driving models or support human-in-the-loop experiments effectively. To bridge this gap, the authors developed JOAN as a Python-based framework that interfaces with CARLA, a popular open-source AV simulation program. JOAN is designed to be flexible, extensible, and accessible, allowing researchers to design, store, and execute human-factors experiments without writing code, primarily through a graphical user interface (GUI). The framework supports a variety of human input devices, including game console controllers, generic USB steering wheels, and high-fidelity haptic devices like the SensoDrive. It is structured into modular components that can be enabled, disabled, or adapted based on specific experimental needs. Standard modules included in JOAN handle hardware inputs, log and plot experiment data, design and execute multi-condition experiments, implement traffic scenarios with dynamic triggers, and provide haptic feedback. The framework’s functionality was validated through its usage in science and education, where it enabled the quick development and execution of several human-in-the-loop experiments, including projects completed by Master’s students within six months. A key feature highlighted is the `SteeringWheelController` module, which supports haptic interaction research—a core focus of the authors’ department. This module allows for the implementation of control algorithms for force feedback, providing example implementations such as a standard PD controller and a Four Design Choice Architecture (FDCA) controller. By leveraging CARLA’s API and pre-trained models while adding a layer of user-friendly experiment management, JOAN democratizes access to high-fidelity simulation tools. The significance of JOAN lies in its ability to lower the barrier to entry for human-AV interaction research. By providing a customizable, low-cost, and code-free environment for setting up complex experiments, it enables broader participation in the development of safe and acceptable AV behaviors. The framework supports reliable data acquisition and repeatable traffic scenarios, making it a valuable tool for both academic research and educational purposes. Its open-source nature and modular design ensure that it can evolve with emerging research needs, facilitating rapid iteration in the study of shared control and human-machine interaction in automated driving contexts.
Key finding
JOAN provides a flexible, user-friendly framework that enables researchers to conduct human-in-the-loop driving experiments in virtual reality without requiring extensive programming knowledge or expensive hardware.
Methodology
other
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.
| Stage | Outcome | Tool | Model | Prompt | Attempts | Completed |
|---|---|---|---|---|---|---|
| discover | success | Crossref | — | — | 1 | 2026-06-05 |
| archive | success | canonical_url | — | — | 1 | 2026-06-06 |
| extract | success | cached | — | — | 3 | 2026-06-10 |
| clean | success | clean | — | — | 1 | 2026-06-07 |
| chunk | success | chunk | — | — | 1 | 2026-06-07 |
| embed | success | embed | Qwen/Qwen3-Embedding-8B | — | 1 | 2026-06-07 |
| enrich | success | openalex | — | — | 3 | 2026-07-02 |
| promote | success | — | — | — | 1 | 2026-06-05 |
| summarize | success | llm | qwen3.6-27b-prismaquant | summ-v5 | 2 | 2026-06-10 |
| tag | success | vector_similarity | — | — | 15 | 2026-06-11 |
| verify | partial | — | — | — | 2 | 2026-06-10 |
Summary generated by qwen3.6-27b-prismaquant on 2026-06-10; 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).
- Methodological Resource: tool software
- Theoretical Contribution: computational model, conceptual framework