Team: 12
School: Multi Schools-EH/NexGen
Area of Science: Mathematics
Proposal: Our project is to implement maze solving algorithms into an autonomous robot which will iteratively improve its efficiency, and compare this to a human’s maze-solving intelligence and learning. This project applies concepts of artificial intelligence, machine-learning, and autonomous robot programming.
In order to solve this problem, we will need to first convert the maze solving algorithms into physical code and apply it to a working robot. We will then wirelessly connect our robot to a data collection system on the computer. Our robot will solve a certain variety of randomly generated mazes and categorize the efficiency (taking elements again from last year’s project). The data will be applied to our robot, which will attempt the most efficient methods first. At the same time, a human will solve the same random mazes. After a certain number of mazes, the speed of completion (efficiency) at the ending maze will be compared between man and machine.
We hope to receive an effective analysis of the efficiency of a simple AI compared to a complex human brain. Not only will this project give us a productive way-finding machine, but a window into the future of artificial intelligence in everyday life.
Sponsor Teacher: Karen Glennon
Mentors: Patty Myers, Neil Haagenson
Team Members:
Quentin Dye
Brendan Kuncel
Savannah Phelps
Sponsoring Teacher: Gary Bodman