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 few 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.
You won't want to miss the 2028 Pan-Asian Deep Learning Conference in Kuala Lumpur:
Call me a hater if you want, but at least for now, pretty sure that's fake. What's funny, though, is that those first five demos are straight out of the standardized humanoid robot demo handbook (which doesn't exist). But seriously: Between ATLAS and ASIMO, walking down stairs, resisting a lateral shove, moving a box, opening a water bottle and pouring water, and kicking a soccer ball is very familiar stuff. In particular, the water bottle opening and soccer ball kicking have been ASIMO standards for years. Even the cart is similar:
[ YouTube ]
Formation-flying drones with fireworks? My New Year's celebrations were nowhere near this cool.
[ Collmot ]
From Torc Robotics:
"Self-driving cars need to be able to handle the myriad complex driving scenarios that happen every day—many of which we cannot predict. Still, we “train” the system to react in a safe way to unpredictable events on the road. This video shows real-world scenarios that our self-driving car has encountered while driving, as captured by the roof-mounted cameras."
"Person unsure of crosswalks." I wonder how much of an edge case that actually is?
[ Torc Robotics ]
We featured one of David Schaefer's Cozmo videos back in November, and we somehow missed the Christmas edition:
That rap song and dance represent 4,000 lines of code. Cozmo may be easier to program than most robots (check out our 2016 interview with Anki co-founder and president Hanns Tappeiner for more), but that's still an enormous amount of work, so here's another New Year's themed video to appreciate:
And just one more, because it has Cinnamon the guinea pig in it:
[ Life with Cozmo ]
In New Zealand, they're letting the dogs sleep in and herding sheep with drones instead.
[ Man and Drone ]
TRI introduces Platform 3.0, the next-generation automated driving research vehicle. The new platform, which is built on a Lexus LS 600hL, combines greater technological capabilities with new harmonized styling that integrates the automated vehicle technology into the LS model’s design.
[ TRI ]
In December, we posted a video from Vincent Berenz at the Max Planck Institute for Intelligent Systems showing an HRI demo with their robot Apollo. This follow-up video shows the demo from Apollo's perspective, via its Asus Xtion.
More details are in the blog post linked below.
[ Vincent Berenz ]
The COBRA-Bee integrates TUI's COBRA carpal-wrist gimbal into a payload for NASA's Astrobee robot. The Astrobee is a free-flying robot that will assist astronauts aboard the International Space Station. The COBRA-Bee will provide the Astrobee with a means of orienting and actuating sensors, grippers, and other tools.
We've asked the folks over at NASA for an update on Astrobee, and we should have that for you in the next few weeks. And if you've never heard of Astrobee before, shame on you, read our article!
Drone-Box is a solution by Delft Dynamics for fast deployment of drones. Due to its automatic door and electric lift, it can be remotely operated. DroneCatcher is a multicopter that is armed with a netgun. It can safely remove unwanted drones from the air. After detection by, say, a radar, vision, or an acoustic system, DroneCatcher is able to quickly approach hovering or moving threats. With the use of multiple onboard sensors, the netgun can be locked on the target. Thanks to DroneCatcher's track & trace capabilities, the drone will be caught by shooting a net. After the catch, DroneCatcher can transport the captured drone (which it hauls via an attached cable) to a harmless place. When the caught drone is too heavy to tow, the target can be dropped with a parachute to ensure low impact on the ground.
While the video seems to suggest that the quadcopter lands back in the box, I'm naturally suspicious, and I'm wondering if it didn't just land behind it.
[ DroneCatcher ]
The TBS Caipirinha II can fly for up to two hours, but we only get four minutes of video to chill with:
[ Team BlackSheep ]
Robotec has a new version of its magnetic guide sensors for automated guided vehicles. Yeah, we know autonomous mobile robots that don't require guides are all the rage, but magnetic tape is cheap, simple, and reliable, and if you don't need versatility, it's not something you need to be embarassed about.
The sensor is primarily used to steer Automatic Guided Vehicles (AGVs), moving material on factory floors. However, its unique sensitivity and accuracy opens a world of new applications such as automatic shelf replenishing in supermarkets, patient transport in hospitals, stage theater props, or rail-less tramways. Compared to other guiding techniques, magnetic guides are totally passive and therefore easy to lay and modify. The tape creates an invisible field that is immune to dirt and unaffected by lighting conditions. The magnetic track can be totally hidden under any non-ferrous flooring material, such as linoleum, tiles, or carpet.
[ Roboteq ]
A guy on YouTube named Linus reviews Jibo, and he's not all that impressed. Except for the twerking. He's impressed with that.
In this week's episode of Robots in Depth, Per interviews Dan Kara, founder of Robotics Trends & RoboBusiness.
Dan Kara talks about how a trip to Japan made him start Robotics Trends & RoboBusiness. He also shares his views on what is going on in robotics. Like many others, Dan found robotics early in science fiction. We hear how the focus has shifted from military applications in the early 2000s to more and more consumer focused progress. Agility in production is also discussed as consumer demand pushes manufacturers to refocus from large scale production of similar items to more customer focused production.
[ Robots in Depth ]