Carnegie Mellon University researchers have built a dataset capturing the interaction of hearing and vision to improve robotic perception and manipulation.

Researchers from Carnegie Mellon University and Facebook AI Research created a navigation system for robots powered by common sense. The technique uses machine learning to teach robots how to recognize objects and understand where they’re likely to be found in house.

Robots typically have trouble grabbing transparent or shiny objects. A new technique developed by Carneige Mellon University hopes to fix that by using color cameras and machine learning to see shapes based on color.

A team from CMU and OSU beat 10 competitors in the Tunnel Circuit stage of the DARPA Subterranean Challenge. The robots conducted a mock search-and-rescue scenario.