Emily Supple ’26 picks up her team’s robot and places it on the corner of a white poster board where a four-by-four grid has been drawn for a game of Breakthrough. The robot — roughly the size of a shoebox — has been fashioned out of white, black, and gray LEGOs. Composed of two rear wheels, an ultrasonic sensor that gives the appearance of two glowing red eyes, and a lifting mechanism similar to that of a forklift’s carriage, the miniature droid’s “brain” is a Raspberry Pi — a series of single-board computers.

On the November morning, Supple turns to her teammate, Todd DiLullo ’26, who’s seated to her right with a laptop, and he runs their Python code. The robot whizzes to life — eagerly moving into the grid.

“On the computer, we have a text file that represents the board,” says Supple, a Marketing major who, along with DiLullo, is working with robotics for the first time. “Zeros represent that the space on the gameboard is empty. One represents the robot move, and two represents the human move. If I save that file and run the prompt through, it should have an updated version of the board, and the robot should now make a move based on where the human went.”

Unfortunately, instead of rotating 90 degrees or moving in a straight line, their robot decides to travel diagonally across the playing field. Supple and DiLullo immediately begin to troubleshoot.

Across the room, Data Science major Hayley Elsbree ’27 is readying her robot for a game of Connect 3. While this is not her first time working with robotics, the experience has been the most challenging — but in the best way.

“You're combining the coding aspect as well as the physical movement of a robot,” says Elsbree. “There are a lot of moving parts to make a robot play a game from scratch.”

Bring on the bots

Led by Biological and Biomedical Sciences Professor Brian Blais, Ph.D., the School of Health and Behavioral Sciences’ “Artificial Intelligence and Robotics” course provides students with an introduction to AI and robotics. Throughout the span of 15 weeks, students create and program their own droids with the goal of developing a robot that can beat a human at a board game of their choosing.

While it’s a daunting task, Blais prepares students for this project on day one — breaking the semester-long assignment into smaller, more digestible pieces. As the project builds, so does the students’ confidence. While not all will go on to a career in programming languages, data collection, or processing analysis, the course teaches valuable skills applicable to other disciplines — such as different ways of thinking and methods for organizing large projects.

Blais notes that students’ shift in confidence typically occurs three-fourths of the way through the semester.

“They start to say and see things in different ways,” says Blais.

To prepare students for their final public presentation in the Unistructure’s Rotunda at the end of the semester, Blais holds several in-class robotics demonstrations where he invites faculty and staff to drop in and ask Blais’s students about their projects. Showing how their bot moves around the game board, they explain how they created their game code, connected their computer to the robot’s brain, taught the robot the rules of the game, and how the robot has a collection of moves it can do in different situations.

“When they finally give a demo of the full game, they see all the pieces come together,” Blais says. “They still have a lot of details to deal with, but they start to see that it works.”

Problem-solving together

Blais’s undergrads have come to find that building robots is an iterative process. Following the November display during the first half of their class session, students spend the latter part further refining their bots.

“When we created our first robot, we showed it to Professor Blais and he said, ‘Good job, now tear it down and make another one,’” reflects Thomas Berger ’25, who — as of early November — was on his third robotic creation. “The first one was pretty good, but it was a little too high and started to break halfway through, so we decided to make one a stabler one that was closer to the ground; a lot of the changes have been efficiency based.”

In addition to solving problems by themselves, the class of 12 congregates to discuss what is and isn’t working for their group. Together they talk through potential solutions — whether it be looking at the differences in power between the robot’s motor and wheels or why basing the robot’s motion on distance rather than time is unreliable.

Blais also inserts practical tips into class conversations, such as debugging the code by breaking it off into a test script.

“Debugging your game in the big file definitely wastes a lot of time,” Blais tells students. “Mostly because it's hard to follow where your robot's going wrong. Be sure to break each behavior into its own test script and, once you've got it fixed, then you move it over.”

Meeting their match

It’s an afternoon in mid-December and six robots and game boards are spread throughout the Rotunda. Seated on one end of the room, Supple and DiLullo are waiting for someone to stop by and challenge their robot to a match of Breakthrough. Since their November demonstration, the two have fixed their directional challenge by speeding up one of the robot’s motors. Their robot also shed several sensors that were causing weight distribution complications.

The use of a camera is also a new addition to their project. Supple explains that the camera would take a picture in the middle of the game and tell the robot what state the board was in. The droid would then make its move based on that information.

“This is definitely one of the most challenging things I've done here — everything else seems easy,” says DiLullo, a Finance major. “I have a new outlook on robotics and a lot of respect for Data Science majors. I’m definitely more interested in doing some of my own projects or looking more into programming’s involvement in finance.”

For Blais, the public demonstration is a day to be proud of.

“When I give them that project description on day one, very few believe that they'll actually get here,” Blais says. “Something may seem insurmountable, but if you break it into lots of little pieces, you can totally do it."