Researchers Validate Key Cutting Control Techniques for Intelligent Deboning System

Research Intern Arnaud Escoffier tests the prototype Intelligent Deboning System, which researchers believe will allow automated deboning systems to match if not exceed the yield and quality performance of the best manual deboning processes.

Research Intern Arnaud Escoffier tests the prototype Intelligent Deboning System, which researchers believe will allow automated deboning systems to match if not exceed the yield and quality performance of the best manual deboning processes.

Researchers at the Georgia Tech Research Institute (GTRI) recently completed preliminary modeling and testing of key cutting control techniques for their Intelligent Deboning System, which they believe will allow automated deboning systems to match if not exceed the yield and quality performance of the best manual deboning processes.

The team is focused on the shoulder cut common to breast deboning. Although automated leg and thigh deboners are in use in many poultry processing plants, automated breast deboning machines are not as prevalent. There are several reasons for this: the natural size variation of birds, the deformable nature of the bird carcass, and variations in the placement of the bird on the cone.

“In order for any system to be a commercial success, it must demonstrate the ability to respond to these product variations as well as meet production and food safety requirements. One production requirement includes the yield of breast meat, while the food safety issue is the presence of bone chips in the meat,” explains Gary McMurray, chief of GTRI’s Food Processing Technology Division and project director.

Here is where the team’s unique cutting control techniques come into play. Much like what the human does when carving a turkey at Thanksgiving, the research team is developing a technique to allow a computer system to guide the blade during the cutting process. The computer is able to cut the meat and tendons, but then automatically guide the blade around the bone without actually cutting into the bone. This technique, explains McMurray, should allow the robot to cut very close to the bone without generating any bone chips – all at line speeds.

Gary McMurray (foreground), project director, and Arnaud Escoffier, research intern, perform testing of the Intelligent Deboning System’s cutting control techniques.

Gary McMurray (foreground), project director, and Arnaud Escoffier, research intern, perform testing of the Intelligent Deboning System’s cutting control techniques.

Initial tests of the prototype system, including cutting experiments, demonstrated the system’s ability to recognize bone during a cut and modify its path to avoid bone chips. While these tests are still very preliminary, the results clearly show the validity of the team’s approach.

The team’s next steps are to refine the prototype to allow more integrated control testing including the use of 3D imaging technology. The 3D imaging is critical to the placement of the blade to start the cut. The imaging algorithms allow the computer to control the blade’s initial position for the unique size and exact position on the cone for each and every bird. Once the cut begins, the control algorithms guide the blade through the shoulder joint and along the scapula.

McMurray says the Intelligent Deboning System could potentially help the poultry industry lower labor costs, achieve higher yields, and improve product safety by reducing the risk of bone or bone chips remaining in the final product. While this technology is still years away from commercialization, researchers are anxious to complete the final system developments so that they can move into more extensive testing of the complete system.


How the Intelligent Deboning System Works

The Intelligent Deboning System or “smart deboner” uses computer vision and other sensing technologies to recognize and react to size and shape differences of a carcass in order to perform precision cuts that optimize yield (the amount of meat removed from the bone) while reducing the risk of bone fragments in finished product.

The system is comprised of a vision system, a cone line, and a cutting system. The vision system identifies the correct starting position for the cut and a nominal cutting trajectory based on the size of the bird. A standard cone line moves the bird through the work cell. The cutting system employs a 2-degree-of-freedom device that is capable of adapting to changes in the internal bird anatomy while compensating for any body deformations.

Thus, as the knife makes strategic cuts on each bird, the computer automatically adjusts the cutting path in response to the particular geometry of the bird coupled with force feedback to perform a near optimal cut.

PoultryTech is published by the Agricultural Technology Research Program,
Food Processing Technology Division
of the Georgia Tech Research Institute.
Agricultural Technology Research Program – GTRI/FPTD, Atlanta, GA 30332-0823
Phone: (404) 894-3412 • FAX: (404) 894-8051
Angela Colar - Editor - angela.colar@gtri.gatech.edu