(Dr. Kok-Meng Lee, advisor)
"Design Algorithm of a Novel Computer-Controlled Gripper for Live Bird Transfer System"
Repetitive handling of natural objects for subsequent meat product processing is often tedious and frustrating. In the poultry industry, the tasks include inserting both legs of a singulated live bird into a shackle at a specified production rate. The objective of this thesis is to develop a simulation algorithm for identifying and analyzing key design parameters that would significantly affect the performance of the gripping process.
The bird is treated as a multi-body system where the legs are modeled as two redundant serial manipulators attached to the birdís torso. The thesis begins with a detailed derivation of the kinematics for the bi-pedal motion. Machine vision and a life-sized bird model are used to verify the kinematics experimentally. The simulation can then be used to analyze the effects of various design and operational parameters on the systemís performance. Attempts will be made to derive an optimal solution within two design constraints. (1) The maximum joint angles of the bird are known and cannot be exceeded without inflicting damage. (2) Joint torques for the birdís legs should be minimized and continuous to eliminate or reduce excess stress. The design algorithm will provide an aid for tuning the design parameters to yield the optimal solution.
The final solution will have an immediate application in the poultry industry. Realistic mathematical simulation and virtual design could reduce construction costs, manufactured hardware configurations, and the quantity of birds used experimentally. It is expected that the analytical method developed here will provide a basis for future related design optimization problems.