At some point in your life, you’ve probably used a combination of sight and touch to find something hidden under the cushions of your sofa. And for a while now, robotics researchers have been trying to give their creations the same ability. Back in 2019, a team of scientists from the Massachusetts Institute of Technology (MIT) used a combination of touch sensors and artificial intelligence to allow a robot to identify objects by touch.
A separate group of scientists at MIT have now built a machine that can find things it initially can’t see. The aptly named RF Grasp relies on a wrist-mounted camera and RF reader to narrow down and pick up an object. As long as an item has an RF tag, the robot can find it, even if it is hidden behind objects like wrapping paper.
The team said MIT News The most difficult aspect of RF Grasp’s development was to integrate both RF vision and RF vision into its decision-making process. They compare the current system to how you might react to sound in the distance by turning your head to locate the source. RF Grasp will initially use its RF reader to find tagged objects, but the closer it gets to something the more it depends on the information it gathers through its camera to make a decision. Compared to a robot with only a visual system, RF Grasp can locate and pick up an object in half the total movements. It also has the unique ability to clean and de-clutter their workspace as they go about their tasks.
Of course, RF Grasp is only better than traditional picking robots at finding hidden objects when tagged RF. But it’s not as much of a barrier as you might think. As MIT News points out, there is already a trend in countries like Japan deploy RF tracking tags in the retail industry. In other words, the technology is already in place for RF Grasp to do its job.