Unboxing the bipedal robot Cassie at the University of Michigan
ฝัง
- เผยแพร่เมื่อ 5 ก.ย. 2017
- A rare bird has landed at U-M: a two-legged robot named “Cassie” that researchers hope could be the forerunner of a machine that one day will aid search-and-rescue efforts, wrote the AP: apnews.com/c57bf857b1e84f7b9b...
Watch more videos from Michigan Engineering: th-cam.com/channels/chi.html...
The University of Michigan College of Engineering is one of the world’s top engineering schools. Michigan Engineering is home to 12 highly-ranked departments, and its research budget is among the largest of any public university.
Built to handle falls, and with two extra motors in each leg, the new robot will help U-M roboticists take independent robotic walking to a whole new level. Follow its progress at the Dynamic Leg Locomotion Lab’s channel: / dynamicleglocomotion
Jessy Grizzle is the director of Michigan Robotics and the Elmer G. Gilbert Distinguished University Professor of Engineering: web.eecs.umich.edu/faculty/gri...
Michigan Robotics: robotics.umich.edu/
Read more
News release: news.engin.umich.edu/2017/09/... - วิทยาศาสตร์และเทคโนโลยี
![[Full Episode] MasterChef Junior Thailand มาสเตอร์เชฟ จูเนียร์ ประเทศไทย Season 3 Episode 1](http://i.ytimg.com/vi/Qs5zdBqLn_8/mqdefault.jpg)
Wooooow. They've been making some good progress.
Usually you hear about technological advancements and then they fall off the edge of the world to never be heard of again.
Needs a sequel (i.e. video showing Cassie's progress).
Cheers to Michigan's Engineering Dept from AREA51.1, somewhere, near Chatham Ontario.
Keep on rockn'
Stability is also a function of the feet the size of the foot is relative to the size of the creature or in this case device. Take a page from a bird and look at their feet. Creating feet which actively respond to the texture of the ground below them can be done with motors of course but some form of smart-micro pneumatic cylinders which are individually or collectively engaged when the pressure on one 'toe' is greater than the others. This also would or could allow the toes to add a sporing motion to each step. Of course adding larger feet means more weight and this requires more power (motors) both for motion and energy to create the motion. I have studied robotic for years and have designed many different kinds. But cost has always kept me from building them. But I still think and dream.
RebelMerc inspiring !!
So when are we going to put a torso, head like Sofia, and dual gun turrets for arms?
I love how there are already sidings with robotics corporations among the civilians.
Like how people side with Boston Dynamics for the backflipping Atlas.
You guys know Boston Dynamics has a robot that does F*****G BACKFLIPS?!?!?!?!
What type of motors does Cassie use?
Add a movable tail for balnce when running or turning around
Cyberball - サイバーボール - Saibābōru is just around the corner...
This robot is using momentum of the legs for staying stable. There is a better solution.
The robots can use a tail to balance the weight, in case it starts to fall. And it can use gyros in every limb, so that they can activate and spin, in case the robot starts to lose balance.
Or the more complex solution would be to use flexible hips, limbs, and limb segments that shift the weight position in every step and wind condition.
Build one then, genius.
He just gave an advice that they could see and implement if they find it worthy. Thank god not everyone is like you Michael, feedback is needed in such projects.
We are one or two innovations away from a rise of truly cool functional synthetic warriors, then we will make a evolutionary step, to replace meat based slobs.:-)
Chicken Legs👍
Yes, shes definitely female.
Since my tax money helped pay for the R&D, do I get stock options in the "spinoff" company? No?
No. Only the wealthy elites get to pillage academia for talent and tech. Everyone else is free to starve; thanks for that seed money, sucker! Ain't capitalism fun? lol
That's not really capitalism. It's fascism, in the truest sense of the word. Socialized risk, privatized profit. It's private corporations getting rich off taxpayer investment. If it was capitalism, the University would also operate as a private corporation on the free market, and not be using government funded labs as their private start-up incubator.
Yeah, kinda. Those with capital bought the lawyers to write the policies and politicians to enact them so that they could bend all of this to their perpetual advantage... Just because your average joe can't read the price tag doesn't mean it isn't for sale. The modern practice of capitalism as it is in reality works a lot like fascism these days, all cloaked from the public.
This is a feeble, diversionary argument. It's death by 1000 cuts
Also $3,460,000,000,000 ($3.46 Trillion; 2017 US tax rev) * 0.000025 (half of a half of 1%) = $86,500,000 (86.5 Million). I'll take some of that. I could do a lot with 1% of that, even in robotics.
Put a fursuit on it, lel
the robot dont work but it have a sticker !!! stupide us
We can't build our own hardware here in the auto-manufacturing Mecca? We need something that looks like a footballers legs that can carry a load. This bird-legged contraption is... Not befitting. Pretty sure birds don't use the staircase to save the burning people.
mistercohaagen Dude it's a locomotion research tool. It's to test control architectures and theories.
Michael Zeagler fair enough, but robotics software can't be abstracted this well yet between platforms... And capitalism dictates that UofM is in competition with everywhere else. I know A2 has the brainstock flowing through it to pull off something greater than we've yet to witness. Most futurists envision Detroit to be the epicenter of the coming revolution... We should start acting like it.
mistercohaagen Capitalism doesn't dictate that at all. Agility robotics is a company in competition, but Jessy Grizzle is essentially a dyed-in-the-wool researcher. This stuff is essentially being open-sourced as it goes. You can look on his website for the Git repos. Also other labs are open-sourcing their work.
And what does that have to do with this robot's legs?
Open Source = awesome; I'm fully behind it. I just see UofM importing hardware from OSU and other places and seriously wonder why we don't have more homegrown hardware. Full disclosure, I'm not being seriously dead critical here... this more of a fan cheering for the home team type b.s. As far as the legs, I saw on another video about the Cassie platform, that they were talking about deploying them to search for the last survivors in a burning building. Along that line of reasoning, it's going to take long enough for an operator to get it up the stairs and around the floors... sending human firefighters in if they do discover someone is going to be way too late. Considering how quickly MI engineering and manufacturing was turned around to serve the war effort in WWII, I don't see why we're on the fence about investing in the technology if lives could be saved (aside from the capitalistic tenancy to ignore the whole of tragedy as "market externalities"; leaving most if not all indefinitely unaddressed).
My actual point is that all city architecture is only built for humans to traverse, so a fully humanoid robot should absolutely be the first milestone we focus on, as it would be the most widely adopted and versatile platform for everything from delivery to search and rescue. "Search" shouldn't come without "rescue" features; they aren't separate things in this context, it's a phrase describing a unified concept (like "space-time"). Unless we get swarming working well, and can build something the size of a golf ball that can navigate a staircase and open doors, a search-only robot is kind of redundant, no? I mean, what would a rescue-only robot look like? It'd probably need search functionality and SLAM crap just to negotiate the environment anyway, so just build one ubiquitous (open source!) platform and use the economies of scale to get this thing out there where it's needed. You could use lighter structure and actuation for software research on smaller, more affordable scales, but the design could then be ramped-up to full-sized humanoid, built from the proper alloys and be immediately useful. Hell, get Honda to open-source their junk and get a running start. I'm sure I'm oversimplifying things as any futurist would, but there is a valid point in there somewhere. We need the right hardware to write the right software, but then why waste effort on these neat-o academic platforms that only seem to produce a multiplicity of novelty, instead of practical functional ubiquity. This is one of the basic ethics of Open-Source.
SHE???? LMAO...
meh, i'll stick with boston dynamics...
I suppose this is like any early stage technology: lots of researchers and ultimately a lot of cross-fertilization, as long as licensing agreements allow. I imagine all the efforts at BD, MIT, Michigan and others are ultimately going to benefit the technology.
Boston DYNAMICS do better!