College students program a robot to take the title of foosball champ.
Computer engineering students at Brigham Young University decided to mix work and play with an experiment that included using artificial intelligence (AI) to predict and adapt to the tabletop game of foosball the same way humans do. Except much faster.
The robotic, vision-controlled foosball table includes software that tracks the ball from a camera overhead, while the computer controls the rods and plastic players. When developing the system, students tried to mimic how humans play the game and programmed that into their code. But the AI algorithms work in real-time, unlike the tired undergrad fueled by chips and pizza. So, guess how wins every time?
The experiment was perhaps too fun, admitted D.J. Lee, an electrical and computer engineering professor, but “through this project students learn the concept of how you can control machines to achieve the tasks that humans can do,” he said in a BYU video.
The camera picks up different colors and the computer calculates in the same way our brains do. AI, however, can predictively adapt to the next move, which makes it difficult for the undergrads to keep up.