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.

Abstract. Automation plays a critical role in the poultry industry, and robotic technology is particularly significant. This paper focuses on developing an autonomous robot system for whole chicken pushing manipulation, a challenging task given the complex, diverse shapes of chickens and their deformable, slippery nature. We explore the use of imitation learning which learns a chicken manipulation policy from human demonstrations. In the experiments, we utilize two imitation learning algorithms, diffusion policy and BC-RNN, on a whole chicken pushing task using the Aloha system as our hardware platform. The results highlight that imitation learning is an effective method for acquiring robust robot policies to manipulate whole chickens. Our research shows that imitation learning could potentially enhance processing efficiency in poultry processing.