Related Links
For the summer of 1996, I had the opportunity to do research in the Robotics and Artificial Intelligence Lab with Maria Gini, a professor that the University of Minnesota at Minneapolis. This was part of the 1996 CRA Mentor Project. In a team of three graduate students and myself, we built mini-mobile robots to participate in the AAAI 96 Mobile Robot Competition that was held in Portland, Oregon. This was an exciting research opportunity for me as an undergraduate. We built these mini-robots out of Legos and radio-controlled car bodies. The main hardware control of the robots came from the use of the Handy Board and the Mini Board that were developed at MIT.
The goal of the competition was to have the robot navigate in a mock office environment given a general map of the floor plan. The robot was to start in an arbitrary office room, and it had to find an empty "conference room" given 3 different choices. Then it had navigate to 3 professor offices and communicate to the professors of the minutes left to the start time of the meeting and the location of the meeting. The robot had to be able to navigate through the hallways and avoid objects and moving people.
The approach we took in dealing with this problem was to use multiple agents working in parallel to get the job done quicker than a single robot. Basically, we created a "Mother Loon" robot that carried on it's back a mini baby robot. The grad students I worked with already had previous experience in building mini-robots using Legos. We were trying to take advantage of smaller size and better speed since we were competing with larger robots that were about 3 feet tall. Check out the official Loon Project page for a more detailed description of the project and for cool pictures of the mini-robots!
Through this project, I had the opportunity to work with software and the hardware, which was exciting given the background of my major. I primarily was responsible for writing the search algorithms and code that the robot used to navigate through the office environment given a general layout of the offices. I also had the opportunity to actually work on building the robots. This was my first hands-on exposure to Robotics and Artificial Intelligence. I enjoyed learning sensors navigation algorithms to use the data from the sensors to perform wall following, corridor/opening detections and the turning of corners.
It was an exciting summer experience because I really got a chance to get exposed to research and the graduate student life. While I was there, I had a chance to learn about other grad students' work in neural networks, computer vision, and decision analysis. From this experience, I have decided that I want to continue on in graduate school in Fall 1997 to get the chance to do more research and study Computer Science in greater depth.
Related Links
People I worked with
