Video Friday is your weekly selection of awesome robotics videos, collected by your Automaton bloggers. We’ll also be posting a weekly calendar of upcoming robotics events for the next two months; here’s what we have so far (send us your events!):
IEEE IRC 2018 – January 31-2, 2018 – Laguna Hills, Calif.
HRI 2018 – March 5-8, 2018 – Chicago, Ill.
Let us know if you have suggestions for next week, and enjoy today’s videos.
PAL Robotics’ StockBot and I share the same holiday party strategy: Find a corner to stand in, and slowly rotate.
[ PAL Robotics ]
At the Autonomous Systems Lab, the Robotic Systems Lab, and the Vision for Robotics Lab at ETH Zurich, sometimes robots do everything to fulfill a child’s dream, even if that dream is of some freaky unicorn monster thing.
Cyclops robot Santa is the BEST Santa.
[ ETH Zurich ]
Happy holidays from UPenn’s resident robots: ESE puppy, Fetch, Spiral Zipper, Crazyflies, Tic Toc, Minitaur, and Turtlebot.
That self-wrapping Minitaur is adorbs.
[ GRASP Lab ]
ArtiMinds Robotics has convinced me that cake should be a holiday staple. And not the gross fruit kind, I mean the kind with frosting.
It’s not every Christmas that you see a Baxter decorating a tree while suffering from a blue screen of death.
[ Dataspeed ]
A feisty happy holidays from Cozmo:
I’m no Santa Claus, but that bot better not be gettin’ between me and my cookies.
[ Anki ]
Here’s a lesson for you, kids: If you asked for a robot for Christmas, the chances are good that it will escape from its box on Christmas eve and hunt you down while you sleep.
[ Reach Robotics ]
Who would have thought that a robot of the Empire might not really get into the spirit of things at Christmas?
[ Sphero ]
Robot company that makes a holiday video without real robots? Happy Holidays ***DENIED***.
[ Fanuc ]
In this video, Cassie Blue fails to throw herself off of a snowy roof:
Must be nice when a falling bipedal robot is just funny and not catastrophic. Also, I heard that "arms are on order" you mentioned in the background... Tell me more!
Paul Ekas from SAKE Robotics noticed some of their dexterous, under actuated parallel grippers in action folding laundry at the iREX trade show:
Turns out the robot doing the folding comes from professor Tetsuya Ogata at Waseda University, who wanted to make sure that we also shared this non-iREX video, showing how the robot performs when it has a bit more time to train using their latest deep-learning algorithms:
You can immediately see why they decided to switch to the SAKE hands: Their compliance allows the robot to pick up fabric on a flat surface, rather than having to cheat a bit with artificial turf like it did before. For more on their robot + AI folding approach, you can read professor Ogata’s paper on the technique in the April 2017 issue of Robotics and Automation Letters.
How to kill your Jibo, should killing your Jibo become necessary.
[ Jibo ]
We wrote about David Zarrouk’s single actuator wave robots last year, but this is a new video of the smallest one yet:
[ David Zarrouk ]
The Clearpath Warthog teleop package makes it easier than ever to send a robot somewhere that you’d never, ever go yourself.
[ Clearpath ]
At the Saint-Gobain plant in Sully-sur-Loire, France, human labor provides high-value work to the finished product. In their shift towards industry 4.0 to free employees from repetitive tasks, Saint-Gobain wanted to automate a grueling glass polishing process, where a complex movement needed to be programmed for every different, small production series of glass.
With the path recording function of the Robotiq FT 300 Force Torque Sensor, the operator can grasp the device and make the movement; the Universal Robots UR10 then records and reproduces the operator’s motion. The operator avoids frequent musculoskeletal disorders and focuses on the glass preparation, a painless, value-added operation.
[ Robotiq ]
Fabian Kung, from the Faculty of Engineering, Multimedia University, Malaysia, wrote in to share a project he’s been working on, building a two wheeled balancing robot equipped with machine vision for navigation: "We are building a bunch of these machines and exploring the possibilities of robot-to-robot coordination so that a group of these machines can patrol and secure a building."
I added a DIY machine vision module (MVM) on my self-balancing robot. The MVM can tilt upwards and downwards along the elevation axis. The MVM complemented the existing infrared sensors on the robot, helping to detect obstacles on the front, sides and also bottom (now the robot can be place on table without the fear of it falling over!). The MVM consists of a CMOS camera connected to a 32-bit micro-controller. The camera captures the scene at a rate of 10 frame-per-second. Each frame is down-sampled to a lower resolution, then subjected to basic pixel/image analysis such as luminescence (brightness) and color extraction, brightness gradient analysis and texture analysis. Based on these and other simple rule-based algorithms, we then decide whether obstacle is present or not. The approach is not 100% reliable, and still needs the support of active infrared sensors in case the machine vision system fails.
[ FK Engineering ]
Even Simone Giertz can’t keep a Kuka collaborative arm from being at least mildly successful at autonomous tree decoration. Mild language warning on this one.
[ Simone Giertz ]
RoboThespian gives a TEDx talk on “Robots, AI, and why the butler didn’t do it.” Or maybe it’s Will Jackson, the founder of Engineered Arts. You’ll never know!
What it means to be artificial and not intelligent in a world where main stream media myths and confusion skew our understandings of robotics and our interpretation of AI. Will Jackson is the director of Engineered Arts Ltd, a robotics company based in Cornwall. Engineered Arts are the market leaders in the design and manufacture of full sized humanoid robots, which are now operational in over 20 countries, with customers including NASA, Science Museum London and Copernicus Science Centre in Warsaw.
[ Engineered Arts ]
In this week’s episode of Robots in Depth, Per interviews Craig Schlenoff, from NIST.
Craig Schlenoff talks about ontologies and the significance of formalized knowledge for agile robotics systems that can quickly and even automatically adapt to new scenarios.
To make robotics systems more agile and easily adaptable to new tasks is very important for robotics to expand beyond large manufacturing settings. Small organizations using robots have new and different needs. They need the robots they use to more easily adapt to their quickly changing needs. Good ontologies and formalized knowledge makes this possible. It might even make it possible to automate the automation.
[ Robots in Depth ]