About the Project

This project involved developing a vision system for a robot arm capable of playing chess. The system was built using Python, TensorFlow, and OpenCV, and it allows the robot to recognize chess pieces, interpret moves, and interact with the board.

The biggest challenge was ensuring accurate piece recognition under various lighting conditions. Through iterative testing and model refinement, the system achieved a high accuracy rate.

Project Overview

This project was part of a second-year group assignment in our university. The goal was to create a robot arm capable of playing chess autonomously. The project was divided among five team members, each responsible for a specific component.

My Role

I was tasked with developing the vision system for the robot. This involved using TensorFlow and OpenCV in Python to enable the robot to visually recognize the chess pieces and their positions on the board. I designed, implemented, and tested the vision software, ensuring it could accurately interpret the game state.

Challenges and Solutions

One of the biggest challenges was ensuring the system's accuracy under varying lighting conditions. By applying advanced image processing techniques and tuning the TensorFlow model, I was able to improve the system's reliability. Training a deep learning TensorFlow model from scratch and collecting enough training data was another significant challenge. This required not only technical expertise but also a strategic approach to data collection and model validation. Regular team meetings and guidance from our supervisor were crucial in overcoming these technical obstacles.

Outcome

Through this project, I deepened my understanding of machine learning and computer vision, and it sparked a keen interest in AI and deep learning. This experience not only enhanced my technical skills but also highlighted the importance of teamwork and interdisciplinary collaboration.