Jesse Hoagg is an Associate Professor in the Department of Mechanical Engineering at the University of Kentucky (UK). He has been with UK since completing his post doctoral studies at the University of Michigan in 2010. His research revolves around the implementation of drones for enhanced human learning. After years of working in aerospace engineering, it might have come as a shock to Jesse that one of his biggest research projects would lead him to become a cowboy.

In the United States, cattle ranches represent the largest of the agricultural sectors. In 2019, agriculture in the US brought in nearly $375 billion. Cattle production alone made up 18% of that revenue with $66 billion. The average cattle ranch is spread over 442 acres of land with about 300 head of cattle. Many ranch owners operate networks of ranches like Stan Kroenke’s WT Waggoner Ranch that has an accumulation of 520,000 acres spread over six counties in Texas. There are approximately 94.8 million cows spread throughout ranches in the US. Every year close to 2.5 million cows die from illnesses or injuries costing the cattle industry close to $1.5 billion.

It falls upon the shoulders of ranch hands, or cowboys, to maintain the health of the cows. It is a nonstop job of traveling hundreds of acres to inspect massive herds, regardless of weather conditions. Finding enough people to fulfill these labor intensive jobs is becoming increasingly harder. As drone technology has progressed, it has made a huge impact in all areas of agriculture. Josh Jackson is a colleague of Professor Hoagg’s and an assistant extension professor in the UK College of Agriculture, Food, and Environment. He also happens to be a ranch owner who has become frustrated with locating his cattle in the dark and often unpredictable Kentucky weather. “Many Kentucky cattle producers have jobs off the farm, and it gets tricky to locate cows this time of the year, when the sun sets so early,” he explained. “We want to lessen producers’ stress by helping them locate their animals quicker and help sick animals faster.”

So he approached Professor Hoagg with a proposal to use drones to help ranchers maintain their herds. The idea would be to use drones to do more than locate and count cattle, but to design a program in which a drone could learn and recognize signifiers of health within the herd. The research project is being backed by the USDA (United States Department of Agriculture) with a $900,000 grant. “This project tackles an important problem, reducing cattle loss,” Professor Hoagg said. “And the approach that we are developing is highly interdisciplinary, drawing on expertise in robotics, computer science, control systems, agricultural engineering, and livestock systems.”

With the help of the USDA, a team of dedicated researchers, and a replica calf named Chuck, Professor Hoagg has been running a series of machine learning and drone formation experiments in the basement of one of UK’s engineering buildings. In the indoor controlled environment, the team is teaching a squad of drones to work together to recognize specific details among a herd of cattle, differentiating between individual cows with facial recognition algorithms. A lead drone locates and oversees the herd as a whole from an elevated position. The lead drone tracks the movement of the herd and compiles a headcount. Then three worker drones can come in closer, yet still at a safe distance, to track specific cows and look for overall signs of health. If one of the worker drones spots anything like lameness, pinkeye, or diminished appetite, that animal can be separated from the herd for treatment.

Professor Hoagg began this project in 2018. Right now they are in the final stages of gathering enough images for the drones to develop a facial recognition database, fine tuning all safety parameters of the fully autonomous system, and testing how cows respond to the presence of drones. To test how the cows respond to the drones, the team has placed heart rate monitors on cows at the nearby C. Oran Little Research Center in Versailles, Kentucky. The drones are flown over the herd 3 days a week, and so far, the monitors have shown no signs of stress in the cows.

It is important that the drones not cause any stress to the cattle, as this could lead to the animals having erratic behavior and decreased appetites. Both of these factors could also lead to increased cattle loss, exactly what the project is trying to prevent. However, the results are looking promising as the cows seem unfazed by the drones, and the system Professor Hoagg is building should be ready by February 2021. “This project aims to make transformational progress on the use of autonomous UAVs for monitoring cattle health and thus improve the security of a critical food resource and improve the economic outlook for rural beef producers,” Professor Hoagg said. If successful, these drones could be of tremendous value to cattle ranchers all over the world.