Author: admin

What started out as a response to labor shortages in poultry processing plants during the COVID-19 pandemic has turned into a robotics system that can learn by imitating human movements to handle chickens.

Using an advanced imitation learning algorithm and camera perceptions, researchers with the Arkansas Agricultural Experiment Station have developed ChicGrasp, a dual-jaw robotic gripper with pinchers that can grasp a chicken carcass by the legs, lift and hang it on a shackle conveyor to be moved on for further processing.

“Embodied AI is used to create intelligent, agent-like robotics to interact with a real-world environment,” said Dongyi Wang, leader of the project and an assistant professor in the Departments of Biological and Agricultural Engineering and Food Science. 

“It’s a physical art that has just developed in the past couple of years, which you see in things like full self-driving cars,” he said. “We are trying to do similar things using that imitation learning idea, but in chicken processing.”

The work has been supported by a $1 million grant from a joint program between the U.S. Department of Agriculture’s National Institute of Food and Agriculture and the National Science Foundation. Wang is a faculty member in the College of Engineering and the Dale Bumpers College of Agricultural, Food and Life Sciences at the U of A. The experiment station is the research arm of the U of A System Division of Agriculture.

Results of the study behind the development of ChicGrasp were published in Advanced Robotics Research. All computer-aided design files, code and datasets from the project were released as open source, providing what the team describes as a reproducible benchmark for agricultural robotics and robot learning.

Imitation learning with ‘diffusion policy’

Traditional robotic methods, such as using suction cups or pre-programmed scripted motions, struggle in the unpredictable conditions of a poultry processing line. The birds are cold, slippery and not uniform in size or posture. Slight changes in leg position or carcass orientation can cause robotics to fail. To address this, Wang’s team designed a system that learns from human teachers rather than treating the gripper and control algorithm separately.

Amirreza Davar, a graduate student in the Departments of Mechanical Engineering and Biological and Agricultural Engineering, designed the gripper and modified the imitation learning to fit into the robotic system, Wang said.

“In imitation learning, the role of the human is to give a trajectory, give a ground truth to the robot, so we don’t need to start from scratch to learn,” Davar said. “It’s more efficient and more accurate. From the get-go, the robot knows what we need to do.”

The camera inputs, movements or trajectories, are stored in a directory that serves as the basis, or “low-dimensional” data, to control each joint in the robotic arm. The specific imitation learning algorithm used, diffusion policy, was introduced in 2023 by Cheng Chi of Columbia University and colleagues at the Toyota Research Institute and the Massachusetts Institute of Technology. 

The system allows for an adaptive framework for continuously refining grasping strategies by formulating robot control as a “conditional denoising process,” Davar explained.

By comparison, other robotics learning methods failed entirely under the same conditions.

“That’s why we’re getting inspired by this algorithm for the poultry industry,” Davar said. “Years ago, robots were programmed specifically to this specific coordinate at this specific time. But what if, like in the poultry industry, things are not predictable? You cannot engineer the robot to go exactly in this position. The chickens come in various sizes, and chicken legs are not always in the same position. So that’s why we wanted the robot to be able to adjust based on that specific scenario.”

Davar said the importance of the work behind ChicGrasp is not limited to the gripper itself. 

“It’s the whole idea of imitation learning and generalization combined with the gripper that makes it applicable and practical in the industry down the line,” he said. 

The speed gap

So far, ChicGrasp has shown a nearly 81% success rate, but the researchers emphasized that speed is still a challenge for industrial use. 

A human can pick up a chicken carcass and hang it on the shackle conveyor in about three seconds. The full cycle for ChicGrasp is about 38 seconds. 

Closing the speed gap will require both motion-level and algorithm-level changes, the study noted. This work would include the use of more aggressive velocity and acceleration limits for the robotic gripper arms and reducing idle time delays.

The cost for building the ChicGrasp prototype, using off-the-shelf robotic arm hardware and 3D-printed pieces for the gripper, was about $59,000.

Open-sourced for the field

By releasing both hardware designs and training data publicly, the team hopes to accelerate innovation in agricultural engineering, where reproducible datasets and benchmarks have historically been limited.

Davar was the first author of the study, titled “ChicGrasp: Imitation-Learning-Based Customized Dual-Jaw Gripper Control for Manipulation of Delicate, Irregular Bio-Products.” Wang is the corresponding author. 

Co-authors included graduate students Siavash Mahmoudi, Chaitanya Pallerla and Pouya Sohrabipour in the Department of Biological and Agricultural Engineering. Pallerla is also in the Department of Food Science. 

Other co-authors included Wan Shou, an assistant professor in the Department of Mechanical Engineering; Phil Crandall, professor of retail food safety in the Department of Food Science; and Zhengtong Xu and Yu She in the School of Industrial Engineering at Purdue University.

This work was supported by awards No. 2023-67021−39072, 2023-67022−39074 and 2023-67022−39075 from the U.S. Department of Agriculture’s National Institute of Food and Agriculture in collaboration with the National Science Foundation through the National Robotics Initiative 3.0.

To learn more about ag and food research in Arkansas, visit aaes.uada.edu. Follow the Arkansas Agricultural Experiment Station on LinkedIn and sign up for our monthly newsletter, the Arkansas Agricultural Research Report. To learn more about the Division of Agriculture, visit uada.edu. To learn about extension programs in Arkansas, contact your local Cooperative Extension Service agent or visit uaex.uada.edu.

About the Division of Agriculture: The University of Arkansas System Division of Agriculture’s mission is to strengthen agriculture, communities, and families by connecting trusted research to the adoption of best practices. Through the Agricultural Experiment Station and the Cooperative Extension Service, the Division of Agriculture conducts research and extension work within the nation’s historic land grant education system. 

The Division of Agriculture is one of 20 entities within the University of Arkansas System. It has offices in all 75 counties in Arkansas and faculty on three system campuses.  

Pursuant to 7 CFR § 15.3, the University of Arkansas System Division of Agriculture offers all its Extension and Research programs and services (including employment) without regard to race, color, sex, national origin, religion, age, disability, marital or veteran status, genetic information, sexual preference, pregnancy or any other legally protected status, and is an equal opportunity institution.