A fundamental part of Olin’s philosophy is that learning is taken out of the realm of academic theory only and put in the context of real-world applications. Students tackle actual engineering challenges in a manner very similar to the global teams assembled by today’s leading companies. This hands-on approach enables them to learn the reality of what it’s like to work within financial or other resource constraints, and how innovative thinking is required to solve real-world problems.
Major Projects
When we first came together as students in the Principles of Engineering class, we knew we wanted to build something that could last beyond our class, was aesthetically pleasing, and that brought joy to people as an interactive object. Soon we realized that all our interested aligned in a project we called Manny the Man (MTM). We decided that Manny would be a sculpture that mimicked the motion of a user. To make our dream for Manny to come true, we broke the idea down into a mechanical, electrical and software pieces. Manny would be controlled with pulleys. He would use an Arduino to control Servos, and it would use a Microsoft Kinect to receive user movements.
After graduating from Centre High School, I wanted to find a way to give back to the school. I combined my knowledge of standarized testing to my web development skills to make an online quiz platform. The program had students create accounts and work through a series of quizzes (50 in total). The quizzes were based on the ACT's mathmaics test patterns.
This project used linear algebra to create, test, analyze, and revise a facial recognition software program. We trained an algorithm on a set uniform images and tested it on random images Principal Component Analysis (PCA). Then we tested all available images to determine an accuracy and efficiency measure. Overall, we found that a voting system among a combination of derivative and eigenface algorithms results in little efficiency. We also found that using the largest variance of the derivative of an image (change in pixels), results in a lower accuracy than the unprocessed image.
