Communications of the ACM
Allen School Releases Robotic Race Car Platform to Drive Advances in AI Research, Education
Researchers at the University of Washington's Paul G. Allen School of Computer Science & Engineering have a platform that can be used to assemble a robotic race car, and also as a tool for experimenting with artificial intelligence.
Credit: Mark Stone/University of Washington
The University of Washington's (UW) Paul G. Allen School of Computer Science & Engineering has released the Multi-agent System for non-Holonomic Racing (MuSHR), a platform for assembling a robotic race car, as a tool for experimenting with artificial intelligence (AI).
The open source, full-stack system can be used to build a race car with commercially available, three-dimensionally-printed components.
MuSHR features computational modules to facilitate intelligent decision-making, along with a software and sensor package to help the vehicle localize in a known environment; a controller can guide the race car toward a goal, while avoiding collisions.
Experimenters can use these capabilities as a base for developing more refined perception and control algorithms.
UW's Siddhartha Srinivasa said, "Beyond research, we...need to think about how we prepare the next generation of scientists and engineers for an AI-driven future."
Srinivasa added, "MuSHR can help...by lowering the barrier for exploration and innovation in the classroom, as well as the lab."
From UW Allen School News
View Full Article
Abstracts Copyright © 2019 SmithBucklin, Washington, DC, USA
No entries found