Wednesday, September 26, 2018

OSCAR Student Michael Norton Researches Turning Ambiguous Traffic Scenarios Into Autonomous Vehicles’ Intelligence

My name is Michael Norton, and I am working on the OSCAR summer impact grant Turning Ambiguous Traffic Scenarios Into Autonomous Vehicles’ Intelligence, which is an interdisciplinary research project between computer science and psychology undergraduates. The main goal of our project is to create an annotated dataset of dash camera footage for objects, action, and lane classifiers to be used by an autonomous vehicle. I first learned about the project last semester through working with Professor Lee, a psychology professor who is heading the project along with Professor Kan (physics) and Professor Lin (computer science). Professor Lee is in the Human Factors department, which focuses on human interaction with technology. I find this to be an important field, as our society’s current explosion of technological innovation requires research on how to design that technology to maximize human usability and minimize potential hazards or health concerns.
In this project, I spend most of my time reading research articles, writing our research paper, and exchanging information and ideas with my co-researchers. Overall, working on this project has been a personally enriching experience for me. I found some of the computer science literature to be impenetrable at first, but learned that persistence and collaboration with those more knowledgeable is a great way to learn difficult new material. I specifically learned a lot regarding datasets, computer vision, and neural networks.  I have also gained greater insight into autonomous vehicle technology, accident analysis techniques, and the overall transportation infrastructure. This project has affirmed my beliefs in the benefits of conducting research for the long term greater good of society.