MIT Creates 3D System That Improves Robots' Spatial Awareness

Robots' capabilities are growing each year, from robot dogs that can do yoga poses to an array of small toy robots to keep engineers entertained, there's something out there for everyone. 

But what about creating an actually useful robot that could help you around the house? 

MIT researchers have been working on robots that they envision to work much like Alexa, if the voice command electronic device was in fact a moving robot. So they have been developing systems to enable these robots to view their physical environment much in the same manner as humans do.

SEE ALSO: FOUR-LEGGED ROBOT CAN 'NINJA WALK' ACROSS NARROW BRIDGES

Robots becoming more and more human-like

"In order to make any decision in the world, you need to have a mental model of the environment around you," said Luca Carlone, assistant professor of aeronautics and astronautics at MIT.

"This is something so effortless for humans. But for robots, it’s a painfully hard problem, where it’s about transforming pixel values that they see through a camera, into an understanding of the world."

Carlone and his team of engineers at MIT have developed a representation of spatial perception so that robots can perceive their environment in a similar way to humans. 

This new model, which the team has called 3D Dynamic Scene Graphs, allows robots to rapidly create a 3D map of their surroundings, including differentiating between different objects such as a chair and a table, as well as humans, rooms, walls, and other structures the robots are likely to encounter. 

The model also enables the robots to question what's in their path, such as the location of an object or the direction of the people in their path. 

Carlone explained "This is not too far from what we do as humans. If you need to plan a path from your home to MIT, you don’t plan every single position you need to take. You just think at the level of streets and landmarks, which helps you plan your route faster."

This new model for robots could not only be useful as home helping robots, but also at higher-level jobs such as working side-by-side with humans in factories, or first responders at disaster sites. 

MIT graduate student and lead author of the study, Antoni Rosinol, stated "With this work, we are making the leap toward a new era of robotic perception called spatial-AI, which is just in its infancy but has great potential in robotics and large-scale virtual and augmented reality."