Sky Chess
A chessboard system that utilizes Machine Learning, OpenCV, and an XY-Gantry to let you play physical chess against a machine. After a quarter, it worked around 65% of the time, and I have plans of making it more reliable.
Skills:
-
Computer Vision (OpenCV)
-
Mechanism Design
-
G-Code
-
Serial Communication
-
API Integration
I created this as apart of an Independent Study in the Spring of my Junior Year; my final presentation for this is at the bottom of the page.
Custom Grippers
To pick up the Chess Pieces, I designed a custom end-effector that took advantage of a Micro Servo and clamps with gears to acquire a piece, and a lead screw with a Stepper Motor to raise and lower the piece.
Computer Vision + AI
To accurately determine the state of the Chessboard, I learned Computer Vision which averages the color values of each square and inputs it into a classifier I trained through sci-kit learn.
Gantry and Communication
For this project I repurposed a cheap laser engraver to serve as my XY-Gantry. The Pi communicated to the Gantry through serial, and pulled from a Python dictionary filled with G-Code positions based on existing logic.
