The Effects of Wing Manipulation on Automated Cutting of Biological Materials

• Main Author: Claffee, Mark Robert
• Published: Georgia Institute of Technology 2007
• Subjects: Ligament; Needle insertion; Puncture
• Online Access: http://hdl.handle.net/1853/16275

Surgical operations and processing of natural product require accurate presentation of the target area in order to achieve precise incisions. An excellent example is the deboning automation for chicken breast meat, for which the pose of the wing can greatly affect the cutting efficiency, ability to fix the structure, and product yield. In contrast to engineering objects, biological products present difficulties such as variation in size, shape, and material properties. Unlike past research, which generally found ways to emulate the manual cutting motion, this thesis investigates the effects of wing manipulation on incision tasks. The objective of this thesis is to develop an analytical model for characterizing the manipulation for pose presentation of a musculoskeletal structure for a specified incision. The manipulation model consists of joint kinematics, the mechanics of bio-materials, and a grasping mechanism to determine the joint pose and forces for a given manipulation trajectory. The model provides a basis for monitoring the cutting of bio-material via non-visual information, as well as for design of a compliant mechanism that can be used in an industrial automation application. To gain a better understanding, a wing manipulation test-bed consisting of a force/torque sensor at the point of wing manipulation has been developed. Two specific examples are investigated. The first is needle insertion into bio-materials, and the other is the shoulder cutting operation associated with chicken breast meat deboning. The effects of manipulation on needle insertion forces are used to quantify improvements in insertion point accuracy and required insertion force. Force signatures are also developed for insertion into the biomaterials located within the shoulder joint. The information gathered from both the manipulation model and needle insertion experiments provide a basis for successful implementation of the automation of the shoulder cut. While the experimentation presented in this thesis is developed in the context of poultry processing, which has immediate contributions as a tool that would facilitate the design of the automated cutting mechanisms in poultry industry, we expect that the development of the models will find a broad range of applications ranging from general meat processing, to surgical simulation, and physical therapy.