Improvement of Driving Simulator Eye Tracking Software
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 report details the development of automated software tools designed to improve the analysis of eye-tracking data collected within the HumanFIRST driving simulator at the University of Minnesota. The research was motivated by the limitations of existing analysis methods, which required researchers to manually review annotated video footage to determine which simulated objects drivers were fixating on. This manual process was identified as time-consuming and resource-intensive. The primary goal was to create a programmatic solution that automatically combines real-world gaze vector data with simulated world coordinates to identify fixation targets without human intervention. The authors developed two complementary software components: an analysis tool and a visualization interface. The analysis software processes inputs from three sources: simulator state data (logged at 60 Hz, including vehicle position and heading), eye-tracking system logs (including filtered gaze headings, fixation IDs, and pupil diameter), and preprocessed metadata defining trial boundaries. The core algorithm determines fixation targets by averaging the gaze heading over a fixation duration and comparing it to the heading of a target vehicle. A fixation is classified as being on the target vehicle if the angular difference is less than 10 degrees, unless the gaze intersects with interior objects like the gauge cluster. The software outputs per-trial summary statistics, including total fixation time, percent time spent on the target vehicle, time to first fixation, and pupillometry metrics comparing baseline pupil diameter to diameter during target fixation. Additionally, a visualization GUI was created to allow researchers to scrub through trial data, viewing the relative positions of the simulation vehicle, target vehicle, and gaze vectors to validate the automated analysis or identify anomalies. The report serves as technical documentation for these tools, outlining the specific input fields required and the output metrics generated. The analysis software successfully automates the extraction of key visual attention metrics, such as the percentage of time spent looking at the road versus interior displays and the specific timing of fixations on target vehicles. The visualization software provides a graphical interface to contextualize these metrics by displaying the spatial relationship between the driver’s gaze and simulated objects. The significance of this work lies in increasing the efficiency and scalability of simulation-based driving studies. By eliminating the need for manual hand-coding of eye-tracking data, the software allows researchers to process large datasets more rapidly and consistently. The tools provide standardized summary statistics and metrics useful across various simulation studies, facilitating more robust analysis of driver visual attention and cognitive load. The report concludes by noting that while the software automates the primary analysis, the visualization tool remains available for cases requiring manual validation or deeper contextual insight.
Key finding
Automated software successfully calculates per-trial visual attention metrics, such as fixation counts and time-on-target, by merging simulator and eye tracking data with a 10-degree gaze threshold.
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. Discovered via bulk_ingest_rosap on 2026-05-23 (6 acquisition events logged).
| Stage | Outcome | Tool | Model | Prompt | Attempts | Completed |
|---|---|---|---|---|---|---|
| discover | success | rosap | — | — | 2 | 2026-05-23 |
| archive | success | — | — | — | 1 | 2026-05-23 |
| extract | success | cached | — | — | 2 | 2026-06-10 |
| clean | success | — | — | — | 1 | 2026-06-01 |
| chunk | success | — | — | — | 1 | 2026-06-01 |
| embed | success | — | — | — | 1 | 2026-06-02 |
| enrich | success | — | — | — | 1 | 2026-05-23 |
| promote | success | — | — | — | 1 | 2026-05-23 |
| summarize | success | llm | qwen3.6-27b-prismaquant | summ-v5 | 3 | 2026-06-10 |
| tag | success | vector_similarity | — | — | 19 | 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.
- gaze based attention detection
- eye movements scanning
- attention allocation
- visual
- useful field of view
- peripheral attention
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, measurement protocol, validation psychometrics