Massachusetts Institute of Technology (MIT) researchers have developed a robot that can sense hidden objects. Called RF-Grasp, it uses penetrative radio frequency to pinpoint items even when they’re not in plain sight.
Radio waves can pass through walls and sense objects hidden behind them. RF-Grasp combines this powerful sensing with more traditional computer vision to locate and grasp items that might otherwise be blocked from view.
Speaking about the project, Professor Fadel Adib, MIT Associate and director of the Signal Kinetics Group, said: “Researchers have been giving robots human-like perception. We’re trying to give robots superhuman perception.”
This could one day streamline e-commerce fulfilment in warehouses or help a machine pluck a screwdriver from a jumbled toolkit.
The research will be presented in May at the IEEE International Conference on Robotics and Automation. The paper’s lead author is Tara Boroushaki, a research assistant in the Signal Kinetics Group at the MIT Media Lab. Her MIT co-authors include Adib, and Alberto Rodriguez, the Class of 1957 Associate Professor in the Department of Mechanical Engineering.
Warehouse work is still usually the domain of humans despite sometimes-dangerous working conditions. That’s in part because robots struggle to locate and grasp objects in such a crowded environment.
“Perception and picking are two roadblocks in the industry today,” says Rodriguez.
Using optical vision alone, robots can’t perceive the presence of an item packed away in a box or hidden behind another object on the shelf. But what visible light waves can’t do – pass through walls – radio waves can.
Like other radio frequency identification systems, RF-Grasp consists of a reader and a tag. The latter is a tiny computer chip that is attached to the object you want to track.
The reader emits a radio frequency signal that gets modulated by the tag and reflected back to the reader, providing information on the tagged object’s identity and location.
With RF-Grasp, both a camera and a radio frequency reader help it find and grab tagged items. It has a robotic arm with a grasping hand and the camera sits on its wrist. It constantly collects both radio frequency tracking data and a visual picture of its surroundings. It then integrates the two data streams into its decision-making process.
“The robot has to decide, at each point in time, which of these streams is more important to think about,” Boroushaki explained. “It’s not just eye-hand coordination, it’s radio frequency-eye-hand coordination. So, the problem gets very complicated.”
Compared to a similar robot equipped with only a camera, RF-Grasp was able to pinpoint and grab target objects with around half as much total movement in a battery of tests. It was also able to declutter its surroundings by removing packing materials and other obstacles to reach the tagged items.
The research is sponsored by the National Science Foundation, NTT DATA, Toppan, Toppan Forms, and the Abdul Latif Jameel Water and Food Systems Lab (J-WAFS).
The research is sponsored by the National Science Foundation, NTT DATA, Toppan, Toppan Forms, and the Abdul Latif Jameel Water and Food Systems Lab (J-WAFS)
