MIT and Princeton's Robo-Picker Can Precisely Lift Nearly Any Object

Thursday, 22 February 2018 - 8:50PM
Technology
Robotics
Thursday, 22 February 2018 - 8:50PM
MIT and Princeton's Robo-Picker Can Precisely Lift Nearly Any Object
< >
YouTube/MIT
As anyone who's ever played one of those claw crane games would know, it's much easier for a robotic claw to drop something than pick it up (especially if, like some of those games, it's rigged to do that)  

Enter the robo-picker machine, created by researchers from MIT and Princeton, which thankfully has much higher ambitions than being some arcade game. As an intelligent robotic claw, it's designed to pick up any household item placed in front of it and determine exactly how to grab or suction it without causing any damage, or disturbing any other items cluttered around it.

It can then decide what the object is and where to place it, making its goals simple, but with an impressive amount of complexity that makes it very good at that simple goal. See it below:



In this case, "impressive amount of complexity" refers to the deep neural network that robo-picker draws from, alongside a system of multiple cameras it uses to scan whatever it's currently holding. This could make it useful in warehouses - Amazon has already shown some interest, as it won an award at the Amazon Robotics Challenge - or in more household chores. Or in an arcade claw game.

According to MIT professor Alberto Rodriguez, who said the following in a press statement:

Opening quote
"This can be applied to warehouse sorting, but also may be used to pick things from your kitchen cabinet or clear debris after an accident. There are many situations where picking technologies could have an impact."
Closing quote




When many similar robots are designed for simple, repetitive tasks, the robo-picker is smart enough that it can diversify what it does (or at least, what it picks up) and that gives it an advantage over simpler factory robots. 

Frankly, unless it is restricted to household chores, it could end up being a player in the continued automation of jobs around the world. But there's still more work that needs to be done on it before that becomes a problem.
Science
Science News
Technology
Robotics
No