Click on “Download PDF” for the PDF version or on the title for the HTML version. If you are not an ASABE member or if your employer has not arranged for access to the full-text, Click here for options. Autonomous robotic system for picking up floor eggs in poultry housesPublished by the American Society of Agricultural and Biological Engineers, St. Joseph, Michigan www.asabe.org Citation: 2017 ASABE Annual International Meeting 1700397.(doi:10.13031/aim.201700397)Authors: Benjamin P Joffe, Colin T Usher Keywords: Agricultural robot, egg picking, poultry, machine learning, TensorFlow, ROS Current conveyor-type systems in broiler-breeder and cage-free layer houses automate transporting of the eggs from the nests to the sorting area. However, the problem of picking floor eggs remains mostly unsolved and requires a farm-hand to walk the house daily, leading to increased labor costs and greater risk of cross-contamination between the facilities. We have developed a robotic system for autonomous egg picking based on a ground rover. The system consists of two parts: the mobile robotic platform with Kinect v2 sensor and deep learning algorithms to detect the eggs on the floor and drive up to them, and a robotic arm combined with visual servoing algorithms to pick up the eggs and place them in a basket. Our approach is designed to be a cost-effective solution that is able to operate in commercial breeder/layer houses and robustly detect eggs in various lighting and environmental conditions. Robust testing of our egg picking algorithms yielded a success rate of 91.6% and we anticipate that the same approach can be used to achieve near-perfect success rate in picking non-damaged eggs. (Download PDF) (Export to EndNotes)
|