Instrumented Farm Vehicle Roadway Study

Self-Propelled Sprayers are driven on roadways from field to field.

Self-propelled sprayers are driven on roadways from field to field.

“This is the first study to use vehicle and roadway GPS and video technology on farm vehicles to develop a safety campaign to improve rural roadway driving behavior at the community-level.” – PI Cara Hamann


Why is this research important?


Farm equipment crashes, which often involve a vehicle rear-ending or passing the farm equipment, are frequent causes of injury.

Occupants in the non-farm vehicle are the most likely to be injured in these frequently severe crashes.

In order to better understand risk factors for rural roadway crashes–especially those involving farm equipment–we have developed a video/GPS system to record vehicle interactions with farm equipment.

Who’s on the research team?


Principal Investigator: Cara Hamann MPH PhD

PI: Cara Hamann MPH PhD

Corinne Peek-Asa PhD

Corinne Peek-Asa

Anton Kruger PhD

Anton Kruger

Natoshia Askelson MPH PhD

Natoshia Askelson MPH PhD

Joseph Cavanaugh PhD

Joseph Cavanaugh PhD

Michelle Reyes, Project Coordinator

Michelle Reyes: Project Coordinator

Research Assistants

  • Nichole Griffith, Undergrad Engineering RA
  • David Wu,  Undergrad Engineering RA
  • Jim Niemeier, Engineering, Assistant Research Scientist
  • Cheryl Roe, Engineering/NADS, Staff Research Assistant
  • Felicia Pieper, Community & Behavioral Health, Research Assistant
  • Whitney Bash-Brooks, Community & Behavioral Health, Graduate Research Assistant
  • Henna McCoy, Injury Prevention Research Center/Epidemiology, Graduate Assistant


What is SaferTrek?


SaferTrek attached to a vehicle.

SaferTrek attached to a vehicle.

SaferTrek is a rear-facing GPS data acquisition system designed to capture the frequency and types of driving behaviors of rear-approaching vehicles. We are examining behaviors such as speed, deceleration while approaching farm equipment, following distance, number of passing attempts, and passing.

What are we doing now?


The most challenging aspect of the image processing has been estimating the distance between the farm equipment and the following vehicle from a single camera view. Three different methods were applied to a validation dataset, and the most accurate method, artificial neural network, will be applied to the roadway data. All three candidate methods require as an input the height of the vehicle captured in the recording.
Therefore, we are currently working to identify the vehicles in each recording so that the distance estimates, which will also be used to estimate relative velocity and classify vehicle trajectories, are as accurate as possible.

GPS Data Processing

Over the last year our team has explored several different approaches for processing the GPS data collected by the SaferTrek devices. New procedures were developed for cleaning and merging two types of raw GPS data and extracting candidate trips. The candidate trips are then analyzed to determine whether the farm equipment traveled on a known road and for what portion of the trip. Various approaches for map matching (i.e., snapping the individual GPS coordinates in a candidate trip to known road segments) and assessing the GPS error, changes in speed and heading, trip length, and data quality have been considered.

Graphical User Interface (GUI) and Coding Protocol for Video Image Annotation

Videos of the vehicle interactions will be annotated to determine when a following vehicle begins to enter the oncoming lane to overtake the farm equipment, as well as passing zones, oncoming traffic,
environmental conditions, and intersections. The codes for annotation were developed and integrated into an existing GUI, which is currently being tested.

Safety Campaign CAB Evaluation

A process evaluation was conducted with the community advisory board who were involved with the development and dissemination of the We’re on this Road Together campaign. Results demonstrated the vital role community engagement played in informing the campaign messaging and coordination of logistics of campaign implementation.

Analysis of Intercept Survey Data

An analysis of baseline pre-intervention intercept survey data was completed during this reporting period and a manuscript is in preparation titled “Predictors of rural driver self-reported passing behaviors when interacting with farm vehicles on the roadway.”

Four data collections (nearly 400 deployments for 3,167 total days) resulted in over 7,000 videos containing more than 2000 on-road vehicle interactions.

What’s next?

We will focus on building the Toolkit, including the focus groups and conjoint analysis. During year two we will pilot the intervention with three communities in Iowa. During year three we will recruit 9 Extension Educators in Indiana, randomly select 4 to be trained and implement the campaign immediately. During year four we will train and implement the campaign with the remaining 5 Extension Educators. During their intervention period, we will provide Extension Educators with technical assistance on campaign implementation and support outcome evaluation data collection. During their inactive year, the research team will use the data collection processes outlined in the Toolkit to collect outcome evaluation data. During year five we will finish outcome evaluation, complete the longitudinal analysis, and disseminate the results of the study. After successfully completing the study, we will have a Toolkit that outlines a process of implementing a community-level intervention targeting safe driving around farm equipment, which could be widely adopted by other states. Additionally, we will have a process for partnering with Extension Educators to implement community-level interventions, which could be expanded to other agricultural health and safety outcomes.


    1. Hamman CJ, Daly E, Schwab-Reese L, Askelson N, Peek-Asa C. Community engagement in the development and implementation of a rural road safety campaign: Steps and lessons learned, Journal of Transport & Health, 2021; 23:101282.
    2. Harland KK, Greenan M, Ramirez M. (2014) Not just a rural occurrence: Differences in agricultural equipment crash characteristics by rural–urban crash site and proximity to town. Accident Analysis and Prevention. 70(8): 8‐13.
    3. Toussaint M, Faust K, Peek-Asa C, Ramirez M.  (2015) Characteristics of farm equipment-related crashes associated with injury in children and adolescents on farm equipment. Journal of Rural Health. DOI:10.111/jrh/1262, Epub ahead of print.
    4. Ramirez M, Bedford R, Wu H, Harland K, Cavanaugh JE, Peek-Asa C (2016). Lighting and marking policies are associated with reduced farm equipment-related crash rates: a policy analysis of nine Midwestern US states. Occupational and Environmental Medicine, Epub ahead of print.
    5. Ranapurwala SI, Mello ER, Ramirez MR (2016). A GIS-based matched case-control study of road characteristics in farm vehicle crashes. Epidemiology, Epub ahead of print.
    6. Swanton AR, Young TL, Peek-Asa C. Characteristics of fatal agricultural injuries by production type. J Agric Saf Health. 2016 Jan;22(1):75-85. doi:10.13031/jash.22.11244. Cited in PubMed;PMCID: PMC5731450
    7. Ramirez M, Bedford R, Wu H, et al. Lighting and marking policies are associated with reduced farm equipment-related crash rates: A policy analysis of nine Midwestern US states. Occup Environ Med. 2016 Sept;73(9):621-6. doi:10.1136/oemed-2016-103672. Cited in PubMed; PMCID: PMC5013097
    8. Ranapurwala SI, Mello ER, Ramirez MR. A GIS-based matched case-control study of road characteristics in farm vehicle crashes. Epidemiology. 2016 Nov;27(6):827-34. doi:10.1097/EDE.0000000000000542. PMCID:
    9. Missikpode C, Peek-Asa C, Wright B, et al. Characteristics of agricultural and occupational injuries by workers’ compensation and other payer sources. Am J Ind Med. 2019;62(11):969-977. doi:10.1002/ajim.23040. Cited in PubMed; PMCID: PMC6944284.
    10. McFalls M, Ramirez M, Harland K, Zhu M, Morris N, Hamann C, Peek-Asa C. Farm vehicle crashes on public roads: Analysis of farm-level factors. J Rural Health. 2021 Sept. Available online 24 Sept 2021. doi: PMCID: NA.
    11. Hamann CJ, Dale E, Schwab-Reese L, Askelson N, Peek-Asa C. Community engagement in the development and implementation of a rural road safety campaign: Steps and lessons learned. J of Tran Health. 2021 Dec:(23). Available online 23 Oct 2021. doi: PMCID: Coming soon.
    12. Arabi S, Sharma A, Reyes M, Hamann C, Peek-Asa C. Farm Vehicle Following Distance Estimation Using Deep Learning and Monocular Cameral Images. Sensors. 2022; 22(7):2736.  (open access)
    13. Ghanbari A, Hamann C, Jansson S, Reyes M, Faust K, Cavanaugh J, Askelson N, Peek-Asa C (2023) Predictors of rural driver self-reported passing behaviors when interacting with farm equipment on the roadway.  Transportation Research Interdisciplinary Perspectives, in press. Paper
Page last updated 09/29/2023