Implementation of Inductive Loop Signature Technology for Vehicle Classification Counts

Minge, Erik; Parikh, Gordon; Dirks, Peter; Jurrens, Chet; Kramer, Isaac · 2024 · ROSA P / Minnesota. Department of Transportation. Office of Research & Innovation

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 evaluates the implementation of inductive loop signature technology for vehicle classification, specifically testing the VSign system developed by CLR Analytics, Inc. The research was motivated by the need to expand vehicle classification data collection beyond limited Automatic Traffic Recorder (ATR) and Weigh-In-Motion (WIM) sites. Traditionally, inductive loop detectors in Minnesota collect only volume and speed data. By analyzing the high-resolution signature produced when a vehicle passes over a loop, this technology aims to provide accurate vehicle classification using existing infrastructure, thereby reducing the time and cost associated with deploying temporary road tubes or piezoelectric sensors. The project team, led by SRF Consulting Group in collaboration with the Minnesota Department of Transportation (MnDOT), installed VSign hardware at five sites across Minnesota: four in the Twin Cities metro area and one in rural northern Minnesota. These sites were selected to represent diverse infrastructure types, including interstate highways, county state-aid highways, and signalized arterials. The team installed detector cards (Phoenix Counter or I-Loop Duo), VSign hubs, and cellular modems into existing cabinets. To validate the system, researchers collected timestamped video footage at each site to create manually verified ground-truth data. This data was analyzed using both the 13-bin Federal Highway Administration (FHWA) and the 7-bin Highway Performance Monitoring System (HPMS) classification schemes. Additionally, the VSign system was compared against the video-based iTHEIA™ system at one location. The results indicated that the VSign system achieved high accuracy for passenger vehicles, with 99% accuracy for Class 2 (passenger cars) and 91% for Class 3 (light-duty trucks) under the FHWA scheme. However, accuracy for heavy vehicles varied significantly depending on the classification scheme. Under the detailed FHWA scheme, accuracy for heavy truck classes ranged from 63% to 87%. Performance improved substantially under the broader HPMS scheme, achieving 81% accuracy for single-unit trucks and 97% for single-trailer trucks. When compared to the iTHEIA™ video system, VSign demonstrated superior performance, achieving 92% classification accuracy versus 86% and a 100% detection rate versus 77%. The study found that system performance was negatively impacted by variations in vehicle speed and lateral position, performing best where vehicles maintained consistent speeds and lane positions. The significance of this research lies in its validation of inductive loop signature technology as a viable method for expanding vehicle classification data collection. The study produced a field deployment manual to facilitate future installations by MnDOT staff. While a preliminary life-cycle cost analysis noted high ongoing software and maintenance costs, the technology offers potential cost savings by eliminating the need for temporary sensor deployment. The findings suggest that while the technology is highly effective for passenger vehicles and performs well under simplified classification schemes, further research is needed to optimize heavy vehicle classification and evaluate cost-effectiveness across different deployment scenarios.

Key finding

The VSign inductive loop signature system achieved high accuracy for passenger vehicles and outperformed a video-based competitor, though heavy vehicle classification accuracy varied and was improved by using simplified classification schemes and maintaining consistent vehicle speeds and lane positions.

Methodology

field_study

Sample size: 5

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.

• naturalistic crash near crash