GIS Mapping of Driving Behavior Based on Naturalistic Driving Data

Balsa-Barreiro, José; Valero-Mora, Pedro M.; Berné-Valero, José L.; Varela-García, Fco-Alberto · 2019 · DOAJ

DOI: 10.3390/ijgi8050226

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 challenge of analyzing massive datasets generated by naturalistic driving studies, which record real-world driving behavior using high-frequency sensors. While these datasets offer significant potential for understanding road safety and traffic dynamics, their sheer volume often forces researchers to thin or reduce data, limiting analysis to specific events or road sections. The authors argue that Geographic Information Systems (GIS) provide a superior alternative to traditional statistical software for visualizing and processing this data, enabling a comprehensive, spatially aware analysis of driving behavior without the need for data reduction. The study utilizes data from the Spanish pilot trial of the PROLOGUE naturalistic driving project, conducted in Valencia in 2010. Five drivers operated a fully monitored vehicle for approximately two hours daily over four days, generating continuous kinematic data (speed, acceleration, braking) and video recordings. The authors developed a specific GIS methodology to map these kinematic parameters along a 15.9 km section of the V-21 motorway. Because raw positioning data are irregularly distributed based on vehicle speed, the method involves creating a theoretical, regularly spaced point array along the road’s central axis. Values from the actual data points are interpolated into this theoretical array using an Inverse Distance Weighting algorithm. The resulting data are then visualized using geometric attributes (polygons projected from the road axis) and color attributes (raster-based color gradients) to represent magnitudes of driving parameters. The results demonstrate that GIS mapping allows for the clear visualization of continuous kinematic variations, such as speed fluctuations, at a microscopic level. The authors illustrate various mapping strategies, including semi-buffers, multiple buffering with isolines, and raster-based color displays, showing how different combinations of geometry and color can highlight specific aspects of driving performance. This approach reveals detailed insights into driver behavior, such as reactions to road geometry (curves, slopes) and traffic conditions, which are often obscured in traditional analyses that focus only on discrete events like crashes. The significance of this work lies in its demonstration that GIS overcomes the limitations of traditional analysis tools, which often suffer from unfriendly interfaces and limited processing capacities for big data. By leveraging GIS, researchers can maintain the integrity of large datasets while gaining a holistic view of driving performance. The authors identify benefits at both macro and micro scales, including the rapid detection of anomalous behavior, the evaluation of road safety infrastructure, and the identification of dangerous road sections. This methodology facilitates multidisciplinary collaboration and provides a more trustworthy, spatially contextualized understanding of driving behavior, suggesting that GIS should be integrated into the standard workflow for naturalistic driving research.

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.

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.

• naturalistic crash near crash

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: dataset resource, tool software