MIT's robot identify objects without even seeing them before

Robots are usually programmed to have the ability to identify objects fed to its algorithms, however, researchers have recently created a new robot that can ‘see’ objects that it hasn’t even seen before.

Researchers from MIT’s Computer Science and Artificial Intelligence Laboratory (CSAIL) have created a computer vision system called ‘Dense Object Nets’ (DON) that can ‘see’ objects it has never seen before.

Computer vision is a machine’s ability to ‘see’ objects or images and understand something about them. DON has the ability to see objects since it looks at them as a collection of points that the robot processes to form a 3D ‘visual roadmap’. Due to this, the scientists don’t necessarily have to sit and label the huge datasets that most computer vision systems require, explained Futurism.

MIT creates blind robot that operates without cameras

When DON sees something that looks like the one it is already familiar with, it is able to identify different parts of the new object. To test this, researchers showed DON a shoe and taught it to pick up the show by its tongue. The system was then able to pick other shoes up by their tongues, even if it has not seen the shoes before or even if they were in various positions than the original one.

The computer system also permits DON to grab a particular object out of a pile of similar objects. “Many approaches to manipulation can’t identify specific parts of an object across the many orientations that object may encounter,” said lead author Lucas Manuelli.

“For example, existing algorithms would be unable to grasp a mug by its handle, especially if the mug could be in multiple orientations, like upright, or on its side,” he continued.

According to Engadget, the system is based on an RGB-D sensor that has a combination RGB-depth camera. The system also is also able to train itself, saving researchers from the hassle of feeding thousands of images of the object to the AI for teaching it.

Though the system is still in its early stages, Manuelli believes that these machines, with their enhanced eyesight and coordination can become a member of people’s household.