Control of an Autonomous Air Hockey Goalie
with Alexander Miller. Completed for University of Florida’s EML 4314C Dynamics and Control Systems Laboratory course in Fall 2015.
Overview
For our final project in UF's laboratory course on Dynamics & Control Systems, our goal was to design and implement a control scheme and strategy for an autonomous air hockey robot. We implemented a homography-based computer vision algorithm in LabView using data from an overhead PS3 Eye camera to track a puck via fiducial markers. A two-axis controllable air hockey paddle system was used, where each axis consisted of one DC motor, a motor driver, an optical encoder, and a belt-drive to guide paddle position along the axis. The two axes were orthogonal and thus independent of one another. A power supply was used to power both motors, and a data acquisition device (DAQ) was used to collect data for use with a LabVIEW VI. A custom air hockey table was used, which was attached to an air supply which pumped air through a series of holes evenly distributed across the tabletop.