AI Conveyor Belt
In this project, my partner and I had to implement an automated conveyor system for sorting electronic components, such as VGA ports, capacitors and resistors, based on their type.
This complex system used a combination of AI-powered vision, precise motor control, and color detection to achieve high accuracy in sorting components into color-coded bins. Notably, we were the only duo to successfully complete this challenge.
Here is a video of our convoyor belt running
We employed multiple IR sensors for the precise detection and positioning of components on the conveyor belt for processing. We also used one RGB color sensor to identify the correct box for each component. This sensor was critical in ensuring that the components were deposited into the appropriately colored box.
Using Arduino microcontrollers, we programmed the stepper motors to control the movement of the conveyor belts. This programming required a deep understanding of microcontroller timers and synchronization, as the belts and sensors had to operate in perfect harmony. To manage the operations of the system, we implemented a state machine model. This model allowed us to control when the belts stopped and started based on inputs from the sensors, ensuring a smooth and reliable sorting process.
One of the most innovative aspects of our project was the integration of vision AI. We utilized a Logitech webcam in conjunction with Google’s Teachable Machine to classify the various electronic components. The AI model was trained specifically to recognize different types of components, enabling the system to make accurate sorting decisions in real-time. This AI-powered classification was a key factor in the success of our project.
Our project not only demonstrated advanced technical skills in multiple disciplines but also underscored the importance of system integration and real-time problem-solving. The successful completion of this project as a duo highlights our strong collaboration, technical expertise, and perseverance.