Scientists recently developed a robot to locate an item that may have been missing for quite some time through the robotic arm attached to it.
A Mail Online report specified that the said robotic arm this model the Massachusetts Institute of Technology researchers developed features a camera, as well as a radio frequency antenna "attached to its gripper."
The invention is working by locating an object through its antenna, bouncing indications off a small, radio-frequency identification or RFID tag that can be attached to an object in the event it goes missing.
The inexpensive and battery-less RFID chips already exist in passports, library books, contactless cards, and the Oyster system, which over 10 million people in London are currently using to pay for public transportation.
How the RFID Tag in the Robot Works
Aside from the public transportation in London, airlines are also using RFID chips for the tracking of luggage and the prevention of shoplifting for retailers.
The RF or radio frequency signals are working perfectly in this situation can travel through the majority of surfaces which include a mound of dirty laundry that may be hiding or concealing missing keys.
Once the antenna has communicated with the RFID tag, which is done by bouncing signals off it in the same way sunlight is reflecting off a mirror, it detects a spherical area in which the tag is found.
Then, the robot combines the said sphere with the input from its camera, which narrows down the location and enables the arm to zero-in on it, move the things above it, and clasp them before confirming that it is picked up appropriately.
A similar TechEBlog report said that according to MIT associate professor Fadel Adib, the notion of locating items in a "chaotic world is an open problem" that has been worked on for years already. He added, having robots that can search for things underneath a pile is a "growing need in the industry today."
The RFusion System
Describing their invention in the MIT report, the researchers said that someday, their RFusion system could sort through piles as well, fulfill orders in a warehouse, determine and have components installed in a car manufacturing plant, or even help a senior to carry out everyday tasks at home.
Adib also explained that at present, one could think of this invention as a "Roomba or steroids," although in the near term, this could lead to many applications in manufacturing and warehouse environments.
To train ta neural linkage that can optimize the trajectory of the robot to the object. In strengthening learning, the algorithm is trained by means of trial and error with a reward system.
That is how the brain is learning, explained the study's co-author Tara Boroushaki adding, one can get rewarded from his teachers, his parents, and computer games, among others.
96-Percent Effective in Retrieving Things
According to the study authors, the robot has a 96-percent effectiveness rate when retrieving completely hidden things under a pile. If one only depends on RF measurements, there will be an outliner, and if he relies on vision, there will be a mistake coming from the camera. However, if they are combined, they will correct each other, according to Boroushaki. And that's what made the RFusion robust.
Information about RFusion is shown on MIT Media Lab's YouTube video below:
Check out more news and information on Robotics in Science Times.