Investigation of Key Safety Measures for Pre and Post-Deployment of Connected and Automated Vehicles
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 study addresses the critical need to accurately evaluate the safety effectiveness of automated vehicles (AVs) in mixed traffic environments. While AVs are designed to reduce driver error, which causes approximately 94% of traffic crashes, their actual safety benefits remain unclear due to limited real-world data. A significant gap in existing research is the use of uniform surrogate safety measures, specifically a 1.5-second time-to-collision (TTC) threshold for both AVs and human-driven vehicles (HDVs). This standard threshold fails to account for the faster reaction capabilities of AVs, potentially misrepresenting their safety performance. The research aims to determine appropriate TTC thresholds for AVs, model their car-following behavior, and assess safety impacts through simulation. The methodology employed a three-phase approach using vehicle trajectory data from the Waymo Open Dataset, focusing on 196 AV-HDV pairs on freeways and arterials. First, the study measured AV reaction times by identifying the lag between speed differences and AV acceleration using cross-correlation and visibility graph algorithms. Second, multiple machine learning models, including Simple RNN, GRU, 1D CNN, and LSTM, were tested to replicate AV car-following behavior. Third, a mixed-traffic simulation framework was developed in VISSIM to investigate safety effects under varying AV penetration rates, utilizing the derived reaction times as critical TTC thresholds. The results indicated that the mean reaction time for AVs is 0.95 seconds on freeways and 1.05 seconds on arterials, significantly lower than the typical human reaction time. When these specific values were applied as critical TTC thresholds, the study found no incidents of potential conflicts for AVs in the dataset, confirming their safety benefits. In modeling AV behavior, the LSTM algorithm outperformed other machine learning models in replicating acceleration patterns. The simulation framework was calibrated to reflect these findings, providing a basis for estimating safety effects in mixed traffic scenarios. The significance of this research lies in its challenge to the conventional 1.5-second TTC threshold, demonstrating that AV-specific thresholds are necessary for accurate safety evaluation. By establishing lower reaction times for AVs, the study provides a more precise method for quantifying their safety advantages. The findings support the potential of AVs to enhance road safety by reducing conflicts, particularly when modeled with accurate behavioral algorithms like LSTM. The proposed simulation framework offers a tool for future research to estimate the safety impacts of varying AV penetration rates, contributing to the development of safer transportation systems and more effective regulatory standards for automated vehicles.
Key finding
Automated vehicles exhibit mean reaction times of 0.95 seconds on freeways and 1.05 seconds on arterials, which, when used as critical time-to-collision thresholds, result in zero potential conflict incidents in the analyzed dataset.
Methodology
mixed_methods
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 | 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.
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: behavioral performance data
- Methodological Resource: tool software
- Theoretical Contribution: computational model