Creating robots that see the world like humans has been a challenge for scientists. While computer vision has come a long way, these systems are still easy to fool. So, why not give robots superhuman perception to compensate? MIT’s Fadel Adib created a robot that uses radar waves to find its target, allowing it to see through walls.
The robot, known as RF-Grasp, has traditional cameras for object recognition. The camera is mounted to the bot’s mechanical grasper, giving it a good view of anything the hand might be trying to pick up. However, what if the target is in a box or under something else? Radio waves can pass through the obstacle, and RF-Grasp can use the reflected signal to spot its target.
To accomplish this, Adib and his team used radio frequency tags, not unlike the ones used to identify pets or open secure doors. The reader sends out RF pings, which power and modulate the tag’s circuits. The reflected signal can provide data, but in this case, it’s being used to track the physical location of the tag.
For the purposes of testing RF-Grasp, the team deployed a small, focused RF reader next to the robot. The reader scans for RF tags in its field of view, and then feeds that data into the robot’s computer vision algorithm. So, when told to pick up an object it cannot see, RF-Grasp relies on the RF pings to seek out the target. When it’s uncovered the object, the robot is smart enough to give more weight to the camera feed in its algorithms. The team says merging data from the camera and RF reader into the bot’s decision-making was the most challenging part.
Compared with robots that only have visual data, RF-Grasp was much more efficient in laboratory tests involving picking and sorting objects. It has the ability to remove clutter from the environment to find its target, guided by RF data that tells it where to dig. For example, it can remove packing material from a box to find something at the bottom. Other robots just don’t have this extra layer of guidance.
This technology could lead to robots that can find objects no matter where they’re hidden. Lost your keys? Just fire up the RF-Grasp Mk V and it’ll figure out which pocket of which coat they’re in. A more realistic application is in the warehouse industry. Robots like Boston Dynamics’ Stretch can pick up and move heavy boxes, but only if they’re visible and regularly shaped. A robot with RF sensing could sort through a messy shelf to find specific objects, not unlike a human. We could be one step closer to eliminating human labor in these environments.
Now read:
- MIT Creates Zoomable Lens Without Any Moving Parts
- New MIT AI Designs Robots On Its Own
- MIT Designs Robot That Eliminates Coronavirus With UV Light
No comments:
Post a Comment