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.

Robotic Rehang with Machine Vision

Published by the American Society of Agricultural and Biological Engineers, St. Joseph, Michigan

Citation:  2021 ASABE Annual International Virtual Meeting  2100519.(doi:10.13031/aim.202100519)
Authors:   Tevon Walker, Konrad Jeffrey Ahlin, Benjamin Peter Joffe
Keywords:   Automation, Deep Learning, Machine Vision, Poultry, Rehang, Robotics.

Abstract. Like many areas in manufacturing, food production is facing serious challenges with sourcing reliable and affordable labor. These challenges are further exacerbated by a rapidly growing demand for food production, labor shortages, and a global pandemic. The process of segmenting a chicken to individual, retail pieces is quite involved, and a key step in this pipeline is placing a processed chicken onto shackles. Traditionally, this is performed by humans; however, the lack of a steady labor supply threatens the throughput of this processing step. In order to ensure food security, current agricultural practices will have no choice but to adopt new technologies and methods, and one such technology is robotics. In this paper, we present an automated system that performs chicken shackling. The system consists of a Universal Robots‘ UR5 manipulator, an Intel Realsense 435d RGB-D camera, and several software components. The camera can perceive a whole, defeathered chicken product in a random initial pose and accurately estimates the position and orientation of the hock. The UR5 manipulator then autonomously grabs this joint with a custom end effector and places it onto a shackle. This technology furthers the automation of the food industry and pushes the application of robotics in a semi-structured and difficult environment.

(Download PDF)    (Export to EndNotes)