Project Summary
This episode of Potentially Genius features an AI-powered system designed to detect whether Brooklyn’s Pulaski Bridge is open or closed. Using an Arduino UNO Q and a USB/Arducam camera, the team collected 73,000+ images of the bridge in different lighting and weather conditions and trained a custom computer-vision model using Arduino App Lab.
The model runs directly on the Arduino through edge AI, while a second Arduino communicates wirelessly to control a miniature bridge model with servo motors and LEDs that mirror the real bridge’s movement.
The project also features a weatherproof camera enclosure made with laser-cut and 3D-printed components, demonstrating a practical combination of AI, embedded computing, wireless communication, and mechanical engineering. Future plans include collecting additional data to predict when the bridge is likely to open.
Github Link to Code: GitHub - dylan-tl/potentially-genius-drawbridge-tracker
Products Used
This project uses several key products to create an AI-powered miniature bridge detection system. The Arduino UNO Q serves as the main computing platform, running the edge-AI computer vision model and controlling the system. An Arducam Camera Module captures images of the bridge so the AI can determine whether it is open or closed. A 9g servo controls the movement of the miniature bridge, while NeoPixels provide programmable LED indicators for visual feedback. A dongle is also used to provide USB connectivity for the camera and other peripherals. Together, these components combine AI, embedded computing, computer vision, and mechanical control into a functional prototype.
YouTube Video Sources
If you would like to watch this video or any others from the Potentially Genius series, please utilize the sources below.
Video on Topic
Arduino UNO Q Bridge Tracker – Potentially Genius® | DigiKey
Other Videos from this Series