Modeling Cooperative Driving Behavior in Freeway Merges
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 challenge of modeling cooperative driving behavior at freeway merges, which are critical sources of traffic bottlenecks. In congested conditions, mainline drivers often cooperate with merging vehicles by decelerating or changing lanes to create gaps, while merging drivers may occasionally force their way in. These interactions mean that driver decisions are influenced by anticipated intentions rather than just immediate situational factors, raising questions about whether merging models developed for one location can be applied to others. The research aimed to develop a utility-based merging model that accounts for unobserved driver heterogeneity and to evaluate its transferability across different freeway networks. The methodology utilized disaggregate trajectory data from the Next Generation Simulation (NGSIM) project, specifically from I-80 and US-101 in California. The model was estimated using maximum likelihood estimation (MLE) to identify parameters most likely to generate the observed vehicle trajectories. To test disaggregate transferability, the researchers employed a likelihood-ratio test comparing “unrestricted” models, where parameters varied between datasets, and “restricted” models, where parameters were assumed identical. For aggregate transferability, the estimated models were implemented in the microscopic simulation tool MITSIMLab. Performance was evaluated using Root Mean Square Error (RMSE) and Root Mean Square Percent Error (RMSPE) to calculate a “Transferability Score,” which compared the accuracy of transferred models against locally re-estimated models. The findings indicated that merging models cannot be directly applied to all congested situations without adjustment. The likelihood-ratio test revealed that applying an estimated model to a network with different characteristics required adjusting at least six parameters, including location-specific constants, standard deviations, and coefficients for driver-specific random terms related to lead and lag gaps. However, at the aggregate level, the study found that after calibration, transferred models could achieve a reasonably close match with local models. The authors noted that aggregate transferability is simulator-dependent, unlike disaggregate transferability. The study concludes that while it provides a rigorous framework for testing the transferability of merging models at both disaggregate and aggregate levels, the empirical case studies are insufficient for generalizing the results. The authors emphasize the need for further empirical studies involving more diverse networks, merging situations, and traffic mixes to validate the findings broadly. This work contributes to the field by highlighting the specific parameter adjustments necessary for model transferability and establishing metrics for evaluating both individual driver behavior and aggregate traffic outcomes in simulation tools.
Key finding
A freeway merging model estimated on one network required re-adjustment of at least six parameters before it could be applied to a network with substantially different characteristics.
Methodology
modeling
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 (8 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 | — | — | — | 4 | 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).
- Theoretical Contribution: computational model