Artificial Intelligence in a competitive robotics competition

Jonathan Krummenacher, Micheal Jones, Angela Rasmussen, Jon Davies, Department of ECE, U of U, 201 Presidents' Cir, Salt Lake City, UT 84112

This is a novel work involving robotics. The topic is Artificial Intelligence in a robotics competition. Our intention is to show our eagerness to master the complex field of mechatronics, and to advance our knowledge of computer vision, simulation of an environment, and AI.

For hardware we are using the Vex platform, and for the competition we are using the RECF's "Change Up". Change Up is a game of 3d tic tac toe, two teams compete against each other during the match. Each team has two robots. Team mates need to work together to manipulate field elements into the scoring towers.

Our Mission with this project is to program an AI to compete in this competition. This involves training the robot in a simulated Gazebo environment running ROS. The robot's are not prebuilt, but need to be designed by the team. The Robot needs to use SLAM, object detection, color detection, sorting of manipulated towers, and doing this in under two minutes. The robot has access to a purtheral of sensors such as Realsense depth camera, IMU's, FLIR black n white camera, and encoders. We also use a Vex Brain as well as a Jetson Nano.

Additional Abstract Information

Presenters: Jonathan Krummenacher, Micheal Jones

Institution: University of Utah

Type: Poster

Subject: Computer Science

Status: Approved

Time and Location

Session: Poster 5
Date/Time: Tue 12:30pm-1:30pm
Session Number: 4058