• Dynamic Video
    Dynamic Video

With the advent of AI, technology has changed drastically. More specifically, machine learning has propelled software forward. Machine learning has given our cars the ability to drive themselves, better spam filters for emails, better text analysis, better and almost real-time speech translations, and even facial recognition in the palm of our hands. As the next generation of software, hardware, and robots is being developed, it is imperative to learn more about how these systems work together. I have decided to create a robotic rover that can use a camera to recognize and label faces based on a convolutional neural network. I gathered multiple face images from my peers and colleagues and used those images to train my convolutional neural network on my computer. This resulted in a matrix of weights, which, when multiplied with the pixel values of an image, could identify the person in the picture. I imported these weights to a raspberry pi, which is a low cost computer and controlled a camera as a webcam. This small webcam is able to recognize faces and then give them a label. Depending on the label, the raspberry pi sends a message to an Arduino. This Arduino, a small microcontroller, controls the motors and display on the rover.