BEHAVE
This research proposal advocates for using imitation learning to train collaborative robots (cobots) for manufacturing tasks, specifically spraying speckle patterns for Digital Image Correlation (DIC).
Summary
The purpose of this effort is to highlight the need for robotic systems that can adapt to unpredictable processes and the limitations of current artificial intelligence (AI) and machine learning (ML) approaches in manufacturing.
Features and Design
Software
Unity
Vuforia
_
Hardware
Android Devices
Tablet and phone
MQ-9 Alpha-Beta Probe
Contributions
Team:
Eli Crawford
Jerry Huggins
_
Clients:
AFLMC/WII
Jamieson Pierce
Outcomes and Additional Information
By using inverse reinforcement learning, the researchers aim to teach cobots (collaborative robots) the necessary skills from human experts, addressing the challenges of workforce aging and knowledge loss.
This project is seen as a valuable step towards a new era of human-cobot collaboration in manufacturing, potentially leading to increased efficiency, improved safety, and a more resilient workforce.
More information and media are coming soon! Thank you for your interest in the project and patience while we add exciting content!