Microscopic Calibration and Validation of Car-Following Models – A Systematic Approach

Treiber, Martin; Kesting, Arne · 2013 · OpenAlex-citations

DOI: 10.1016/j.sbspro.2013.05.050

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 addresses the critical need for systematic calibration and validation techniques for microscopic car-following models, which are essential for accurately describing driving behavior and traffic flow dynamics. The authors argue that default parameter sets are insufficient due to variations in driver styles, vehicle types, and traffic regulations across different regions and times. Consequently, they investigate key unresolved questions regarding data requirements, the influence of sampling rates and measurement errors, the impact of data smoothing, and the comparative performance of different calibration methods. The study aims to establish a rigorous framework for assessing model robustness, parameter orthogonality, and fitting quality. The methodology employs both real-world and generated data to isolate influencing factors. Real data includes extended floating-car data (xFCD) collected from instrumented vehicles in German inner-city streets and trajectory data from the NGSIM project. The authors emphasize the necessity of preprocessing data to ensure internal and platoon consistency, deriving positions and accelerations from primary measurements like speed and gap to eliminate redundancies and errors. Two primary calibration approaches are compared: a local maximum-likelihood method, which compares model-predicted acceleration against observed acceleration at each time step, and a global least-squared errors (LSE) approach, which minimizes the sum of squared errors between simulated and observed trajectories. The global approach utilizes objective functions based on absolute and relative gap differences, with the latter shown to be less sensitive to outliers. The Intelligent Driver Model (IDM) is used as the primary test case, alongside the Optimal Velocity Model and Full-Velocity Difference Model. Key findings reveal significant differences between calibration methods. Local maximum-likelihood calibration often yields unrealistic parameter values, such as an unrealistically low maximum acceleration, and produces serially correlated errors that invalidate standard statistical error estimates. In contrast, global LSE calibration using relative gap differences provides more plausible parameter values and better handles various traffic regimes. The study demonstrates that acceleration-based objective functions are insensitive to certain parameters, such as desired speed, while speed-based functions ignore gap-related parameters. Furthermore, the authors highlight that data noise and low sampling rates significantly bias calibration results, particularly for less robust models. Generated data allowed the authors to quantify these biases, showing that smoothing and resampling must be carefully managed to maintain data consistency. The significance of this work lies in providing a systematic approach to microscopic model calibration, moving beyond simple fitting quality to consider robustness and parameter plausibility. The authors conclude that global calibration based on relative gap errors is superior for ensuring consistent and realistic parameter estimation. This framework helps distinguish between model deficiencies and data artifacts, offering guidelines for future research on data requirements and the separation of intra-driver and inter-driver variations. By establishing these rigorous standards, the paper enhances the predictive power and applicability of car-following models in traffic simulation and analysis.

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.

StageOutcomeToolModelPromptAttemptsCompleted
discover success OpenAlex-citations 1 2026-06-19
archive success openalex 5 2026-06-25
extract success cached 2 2026-06-26
clean success clean 1 2026-06-19
chunk success chunk 1 2026-06-19
embed success embed Qwen/Qwen3-Embedding-8B 1 2026-06-19
promote success 1 2026-06-19
summarize success llm qwen3.6-27b-prismaquant summ-v5 1 2026-06-26
tag success vector_similarity 6 2026-06-19
verify success 1 2026-06-26

Summary generated by qwen3.6-27b-prismaquant on 2026-06-26; 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).