Principal investigator: Ilya Kovalenko
University: Pennsylvania State University
Industry partner: DMI Companies, Inc.
Robotic systems can enhance efficiency, reduce costs, and boost productivity for manufacturers. However, current robotic solutions are typically limited to pre-programmed tasks; adapting robots for new tasks often requires specialized expertise. This research aims to increase robotic adaptability for material handling in manufacturing, where requirements may shift due to changes in product or process specifications. The proposed approach focuses on developing a large language model-based framework for adaptive and collaborative robots. Specifically, the project will focus on developing an adaptable vision system and a dynamic path-planning algorithm that integrates feedback from an operator on the shop floor to improve robot performance. This work will improve the reusability and adaptability of robots for small and medium manufacturers (SMMs), allowing verbal operator feedback to modify robot capabilities. The developed framework will be tested at the DMI manufacturing facility, where the robot will need to adapt to different material handling tasks and varying product dimensions.