Developer - Robot Learning Lab
"Cloth Grasp Point Detection based on Multiple-View Geometric Cues
with Application to Robotic Towel Folding"
by Jeremy Maitin-Shepard, Marco Cusumano-Towner, Jinna Lei and Pieter Abbeel
Printable View
Developer - Robot Learning Lab
"Cloth Grasp Point Detection based on Multiple-View Geometric Cues
with Application to Robotic Towel Folding"
by Jeremy Maitin-Shepard, Marco Cusumano-Towner, Jinna Lei and Pieter Abbeel
https://youtu.be/gy5g33S0Gzo
(50X) Autonomously folding a pile of 5 previously-unseen towels
Uploaded on Mar 17, 2010
Quote:
Project was led by UC Berkeley Ph.D. student Jeremy Maitin-Shepard working with Professor Pieter Abbeel.
https://youtu.be/-lBHNbiooxQ
(30X) Autonomously folding a towel from an unusually challenging initial configuration
Uploaded on Mar 17, 2010
https://youtu.be/Thpjk69h9P8
Laundry-Folding Robot from UC Berkeley
Uploaded on Jul 12, 2011
Quote:
UC Berkeley PR2 Robot Folds Laundry
In 2010, the Berkeley PR2 (unofficially named Brett, for "Berkeley Robot for the Elimination of Tedious Tasks") showed us that it could pick a towel out of a pile of clean laundry one by one and neatly fold and stack them, which was an impressive demo. Somewhat less impressive was the fact that the robot would take between 20 and 25 minutes to neatly fold one single towel, which, let's face it, isn't entirely practical. That time has now been cut down to under six minutes, with the potential for as little as two minutes per towel if they really crank the robot up.
The key to robots is in advances in Artificial Intelligence (AI). A limiting factor in developing a useful and practical robot is that current robots can only do what their programming tells them. You can program a robot to fold a shirt, but if you throw in a shirt with buttons on the opposite side, the robot may not be able to fold the shirt. If you program a robot to bake a cake, there is little chance the robot can make lasagna without specific programming.
The PR2 doesn't rely on such discreet programming. Rather, it reduces the dependence on a complex vision system and instead relying on gravity and the properties of cloth. The PR2 now just picks up a towel wherever is convenient and then drags it across a folding table, knowing that as it does, the piece of the towel furthest away from the gripping point must necessarily be a corner.
What do you all think? Is this a big step forward in robotics technology or just another robot marginally better than the last?
Professor Pieter Abbeel is optimistic that by the time the PR2 Beta Program ends in 2012, they'll be able to do an entire laundry cycle from start to finish.
https://youtu.be/IBY4t8XxH7E
Robot tying knots and folding towels - What UC Berkeley and Intel are up to in Robotics
Published on Jun 24, 2013
Quote:
Robot tying knots and folding towels - What UC Berkeley and Intel are up to in Robotics - We visited the Robotics lab at UC Berkeley to see what Intel and their partners at the CITRIS robotics lab in Berkeley where up to. In this video you see a Willow Garage PR2 robot, that has been trained to fold laundry and tie knots all by itself without any outside intervention. It is one of the first steps towards robots that'll soon help us get simple house work like laundry done and are able to assist in tasks in surgery like suturing, where a robot will be able to work without the tremors and other issues that human surgeons often have to deal with while working on patients.
https://youtu.be/N_co3vimPR4
Autonomous robot doing laundry
Published on Nov 19, 2014
Quote:
Given only the high-level goal of cleaning all "dirty" clothes (which are unknown in number) and actions such as pick, place, open and close, the UC Berkeley PR2 figures out a plan for doing laundry and then executes it.
https://youtu.be/fheA-lgVaxc
Robotic Towel Folding -- Vintage Edition
Published on Apr 7, 2016
Quote:
This is a video of our first successful towel folding effort back in summer 2009. After this result we transitioned to working with Willow Garage's PR2.
Berkeley Artificial Intelligence Research lab (BAIR) -- bair.berkeley.edu
This work was led by PhD student Jeremy Maitin-Shepard, in collaboration with undergraduates Marco Cusumano-Towner and Jinna Lei, and professor Pieter Abbeel.