Description

Farm equipment with embedded algorithm and camera system detects, notifies unsafe driver behavior

The Ag-OM system identifies risky behavior from farm equipment operators, alerts them, and prompts behavior adjustment using a machine-learning based algorithm. Tractor operators are more likely to sustain falling injuries depending on how they enter and exit the tractor, whether they face inside or outside the cab, and if they utilize the hands and steps for maximum safety. Reducing these types of falling injuries could reduce some of the 20,000 agricultural falls that occurred between 2015 and 2019. This trained algorithm uses cameras and an open-source, human-pose-detecting algorithm to monitor driver safety during these specific times.
 
There is a limited amount of safety equipment and alert systems that monitor preliminary stages of machine operation like entering a tractor cab. Instead of focusing on the machinery, this human-centered approach provides real time alerts and noises, and the driver can change their behavior instantly.
 
In the future, this platform can be used to identify other instances in which human pose detection could signal unsafe working conditions. How operators enter and exit the cab is the first of many points along the operation pathway that could be monitored to reduce injury.
 
To discuss licensing opportunities contact Tyler Scherr, PhD, at tyler.scherr@unmc.edu or 402-889-5498.