 |
| RF-Grasp ... a hidden object-sensing robot |
Researchers at the Massachusetts Institute of Technology have developed a robot called RF-Grasp that uses radio waves that can pass through walls to detect things that you cannot see at first.
RF-Grasp combines sensors with traditional image processing to identify and detect potentially invisible objects.
And developments may one day simplify warehouse e-commerce tasks or help machines remove screwdrivers from mixed toolboxes.
Fadel Adeeb, an assistant professor at the Massachusetts Institute of Technology, said: Researchers give robots a human-like perception, and we're trying to give robots a supernatural vision.
With the continuous development of e-commerce, despite the sometimes dangerous working conditions in general, warehouse work is still an area of human labor, not a robot.
One of the reasons for this is that it is so difficult for robots to locate and sense things in a crowded environment as perception and choice have become obstacles in today's industry.
With eyesight alone, the robot cannot see the objects hidden in a box or behind another object on the shelf, because, unlike radio waves, visible light does not pass through walls.
Even when the camera is completely hidden from the camera’s field of view, RF Grasp uses both the camera and a radio frequency reader to search for and capture special things.
As long as the object has the RF tag, the robot can find it, even if it is hidden behind an object like wrapping paper.
Unlike typical university robotics projects, the present invention also has unique use cases. The team believes RF Grasp helps companies like Amazon improve and simplify warehouse automation.
The most difficult aspect of developing RF Grasp was the incorporation of image and radio frequencies into the decision-making process. The team compared the current system to how you reacted to distant noises as you turn your head to determine the source of the sound.
RF Grasp uses an RF radio frequency reader to find unique objects. The closer it is to the object, the more it relies on information gathered by the camera to make decisions.
Compared to a robot that only operates on a vision system, the RF Grasp robot can locate and capture objects in about half of the total movement and has the unique ability to clean its working area while performing tasks.
Additionally, there is a trend in Japan and other countries / regions to develop radio frequency tracking beacons in retail, which means that the technology is already in the works to make the RF Grasp work properly.