Two-Dimensional Modeling of Bicycle Behavior
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 addresses the lack of comprehensive models for bicycle traffic flow dynamics, specifically the absence of frameworks that capture the non-lane-based, lateral movement characteristics of cyclists. While vehicular traffic modeling is well-established, bicycle research has historically focused on longitudinal motion or mixed-traffic interactions, often relying on simplified assumptions that ignore the flexibility of bicycle lateral maneuvers. The study aims to develop a descriptive, two-dimensional model that accounts for both longitudinal and lateral behaviors, including overtaking, while remaining sensitive to cyclist physical characteristics, bicycle conditions, and roadway surfaces. The methodology expands upon the existing Fadhloun-Rakha (FR) bicycle-following model, which governs longitudinal motion. The researchers complemented this with a novel lateral motion strategy based on point-mass dynamics. This strategy utilizes three predefined elliptic regions around each bicycle: a view zone, a steady-state zone, and a safety zone. The model determines angular motion and overtaking feasibility by analyzing the intersection of the steady-state zone of a following bicycle with the safety zone of its leader. This approach allows for dynamic, stochastic lateral movements that mimic naturalistic flock-like behavior. To validate the model and support future applications, the team also developed a framework for extracting naturalistic cycling data from video feeds. Using computer vision and machine learning techniques—including HOG detectors and semantic segmentation—they processed continuous video data from a non-signalized intersection at Virginia Tech. This process involved trajectory extraction, intersection surveying, and coordinate transformation to determine precise location, speed, and acceleration profiles. The study successfully produced a two-dimensional model that is the first of its kind to describe bicycle longitudinal and lateral behavior in both constrained and unconstrained conditions using point-mass dynamics. The model integrates the FR longitudinal logic with the new lateral module, enabling it to simulate overtaking maneuvers and lateral variability. The data collection effort resulted in a dataset of 619 high-precision bicycle trajectories. The extracted data provided detailed profiles of distance traveled, speed, and acceleration, confirming the feasibility of using computer vision to generate naturalistic cycling datasets where traditional instrumentation is impractical. The significance of this work lies in providing a robust, physics-based tool for simulating bicycle traffic that accounts for individual cyclist variability and environmental factors, unlike previous models that relied on generic vehicular analogies. The developed model offers a necessary foundation for more accurate traffic simulation and planning for bicycle infrastructure. Additionally, the creation of the naturalistic dataset addresses a critical gap in the availability of empirical cycling data, offering a valuable resource for the research community to further develop mobility applications and traffic safety analyses.
Key finding
The proposed two-dimensional model successfully integrates longitudinal and lateral bicycle motion dynamics, validated through the extraction of 619 naturalistic trajectories using computer vision techniques.
Methodology
mixed_methods
Sample size: 619
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 | — | — | 24 | 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).
- Theoretical Contribution: computational model