AIMM ICC

I was one of the team members to represent Purdue University in the Artificial Intelligence Maritime Maneuver Indiana Collegiate Challenge. Our objective was to design and build an autonomous boat equipped with cameras for object detection to identify the colors of obstacles. The competition featured a racetrack set up on a lake, and the boat had to perform specific movements based on the obstacle’s color detected by the cameras. Points were awarded for correctly executing these maneuvers, such as moving left to pass through a green obstacle or moving right to pass through a red obstacle. Additional bonus points were given if the boat successfully completed the track and returned to its starting point.

I was part of the computer vision team, responsible for designing the data pipeline for the cameras to perform obstacle detection. The other teams included the mechanical team, which handled the physical mountings on the boat and ensured all systems had power, and the navigation team, which used the computer vision detections (including distance data) and GPS coordinates for navigation.

We published a technical report detailing our process as part of the judging for the AIMM competition. Unfortunately, due to a late start and extremely bad weather, Purdue’s performance in the competition was not as strong as we had hoped. Despite these challenges, we successfully developed a fully remote-controlled boat with a working computer vision system capable of detecting obstacles and calculating their GPS coordinates based on their distance from the camera for navigation purposes.

The technical report for the competition can be found here.