The robot revolution is headed for the sheep pen.
Australian Wool Innovation (AWI), a not-for-profit R&D firm, has linked arms with NextGen Agri and the University of Sydney on a first-of-its-kind investigation into the potential of artificial intelligence (AI) in the sheep and wool industry.
The first stage of the project is to demonstrate how novel phenotyping technologies such as image analysis, biomarkers and bio-sensing, combined with deep learning, can “unlock new horizons for the Australian sheep industry,” said Jane Littlejohn, general manager of research at AWI.
“This project will provide a proof of concept that semi-automated images captured and combined with machine-learning techniques can be used to determine identification through facial recognition, wrinkle scores, face cover and live weight in sheep,” Littlejohn explained.
The project’s long-term goal, she added, is to evaluate the use of advanced phenotype and AI technologies to predict—and manage—a sheep’s lifetime performance from a young age. It could assist breeders in their selection process by predicting, with a high level of accuracy, how certain combinations of genes manifest complex physical traits.
The results of the investigation could also inform AWI’s farm automation program, which aims to make remote properties “fully serviced” with WiFi and smart tags, said Mark Ferguson, project lead at NextGen Agri, a livestock biotechnology firm based in New Zealand.
“This project will also lay the foundation for new and innovative ways to assess traits in sheep without additional time and effort from farm managers,” Ferguson said.
Animals, for example, might be automatically weighed and identified without “extensive infrastructure.” Even wool shearing might someday be fully automated using modular, portable robotic sheds.
Related Stories
“Our long-term aim is to develop a readily available and capable automated alternative to manual shearing that will provide a range of benefits including ensuring the welfare of the sheep and reduces the risk of both human and animal injury,” said Richard Lyons, project director at Robo Shear, which is working with AWI and Ranken Research on a four-year initiative to mechanize wool harvesting.
Such a system, Littlejohn said, could help fill a manpower shortage in Australia, where the number of shearers has plunged 13 percent in five years to just 2,482, according to a 2016 report from the Australian Bureau of Statistics.
But the project will only be hailed a success if the robots can measure up to humans in speed and quality.
“It is critical the end product of our project ensures the quality of the fleece with a target rate of 1,500 de-fleeced sheep in a continuous 10-hour period,” Lyons said.