Development of the ConnDOT horizontal curve classification software.

Osei-Asamoah, Abigail; Jackson, Eric · 2014 · ROSA P / Connecticut. Dept. of Transportation

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 report details the development of automated software to classify horizontal curves and grades for the Connecticut Department of Transportation (ConnDOT) to facilitate reporting to the Federal Highway Administration’s Highway Performance Monitoring System (HPMS). Manual updates for Connecticut’s over 2,000 HPMS sites are time-consuming and prone to error. The research aimed to create a user-friendly, batch-processing tool that extracts geometric data from existing inventory datasets to generate accurate curve classifications, thereby improving efficiency and data reliability. The methodology utilized data collected by ConnDOT’s Automatic Road Analyzer (ARAN) van, which records GPS location, gyroscope heading, and grade data at four-meter intervals. The researchers processed this data using MATLAB, applying smoothing techniques—specifically moving average and Savitzky-Golay filters—to remove noise and outliers. Two classification approaches were investigated: a "Per Point" method, which classifies every data point based on delta heading calculations, and a "Per Curve/Tangent" method, which identifies specific curve start and end points using a heading change threshold. The Per Point method was selected for final implementation because the Per Curve/Tangent method failed to detect gentle curves on interstates due to gyroscope precision limits. Curves were categorized into six classes (A–F) based on degree of curvature, and grades were classified by percentage, adhering to HPMS standards. Validation was conducted by comparing software outputs against design drawings for specific sections of Route 44 and Route 20. The Per Point method accurately identified Curve Type D for Route 44 sections and Curve Type A for Route 20, with minor discrepancies attributed to vehicle trajectory variations and instrument precision. The researchers also implemented error-checking routines to correct anomalies caused by ARAN van lane changes and turning maneuvers at route endpoints. The final output consists of executable files that generate curve classification reports and calculate the lengths of specific curve and grade types for any given roadway section. The significance of this work lies in the creation of a robust, automated system that replaces inefficient manual processes for HPMS reporting. By leveraging existing photolog survey data, the software provides ConnDOT with a scalable solution to maintain accurate geometric inventories. This tool not only saves significant time and resources but also enhances the precision of data submitted to the FHWA, supporting better highway planning, safety research, and federal fund appropriation. The report concludes by suggesting future studies to further refine the classification algorithms and expand their applicability.

Key finding

The per-point classification method accurately identified horizontal curve categories by calculating degree of curvature from ARAN van heading data, whereas the per-curve/tangent method failed to detect gentle curves on interstates due to gyroscope precision limits.

Methodology

dataset

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).

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.

• road geometry