Poster Session 4, 2:15 PM - 3:00 PM: Campus Center Auditorium [A27]

Training a Robotic Arm With AI

Presenter: Caileen Joan Wilson

Faculty Sponsor: Hong Yu

School: Fitchburg State University

Research Area: Electrical and Computer Engineering

ABSTRACT

In the last decade, advances in robotics using artificial intelligence (AI) are starting to make human life easier in both industrial and educational settings. These robots are being trained and used to carry out specific tasks, such as picking up and recognizing objects, and movement. In order to carry out these tasks, they rely on different sensors, such as light, color, and touch sensors. In this project, the main focus is getting the robot to recognize different colored objects, sort them, and implement force-controlled pick-and-place for the colored objects. This shall be executed by programming a Robotic Operating System (ROS) into a Quanser QArm. The QArm is a robotic arm apparatus. It works with MATLAB with Simulink and Python as an implementation platform, along with their add-on QUARC. The building platform can be used to program the ROS. Students can also use their platform for research in Artificial Intelligence (AI) and Machine Learning (ML). This programming can be used with the Quanser QArm to train it to perform specific tasks. These tasks include but are not limited to picking up and putting down objects, and recognizing and sorting objects by other features. The goal is to gain an understanding of how Artificial Intelligence and Machine Learning principles can be applied to servicing industrial and educational settings, as well as demonstrating this understanding by training the Quanser QArm to perform a series of tasks. 




RELATED ABSTRACTS