DARPA's next Grand Challenge to focus on humanoid robots | Hacker News Reader

DARPA's next Grand Challenge to focus on humanoid robots

engadget.com

50 pointsspathak26 comments

Hey spathak... I'm the author of the original source over at Hizook.com (which was referenced in the Engadget article):

http://www.hizook.com/blog/2012/04/03/new-darpa-grand-challe...

We work very hard to put together these top-notch articles, and it sucks to have other media sites get all the attention.

EDIT: I had already made a HN submission for Hizook, here: http://news.ycombinator.com/item?id=3806243 but I suppose this post already has a lot of comments, so it should (obviously) remain. However... if you're interested in this type of robotics news, be sure to check us out.

Not to be a downer but if you want to be taken seriously for reporting your url probably shouldn't be [silly sounding domain]/blog. It doesn't lend to credibility and I'm sure I'm not the only one that given two similar titles but one has (hizook.com) next to it will click the one with the more serious sounding domain.

Hizook is no more "silly sounding" than many other household-name web domains that you probably use on a daily basis (and it has special meaning to me). The reason the articles are at /blog is because I'm developing hardware products for the site. (I'm not a "journalist", I'm a hardware hacker with a PhD in robotics.)

As to credibility... Hizook blows Engadget out of the water. We're already well-known by most academic and professional roboticists, and the discussions on the site do not revolve around "robot overlords." We've been called "one of the top robotics blogs" by IEEE Spectrum, where I'm also a contributing author. Oh, and the HN guidelines specifically state to: "cite the original sources."

Respectfully, your feedback is trite. I suggest checking out our site's articles, and if you'd like to continue this discourse, then please feel free to email me (see HN profile).

but if you want to be taken seriously for reporting your url probably shouldn't be

What the hell?

Are we judging content by URLs now? And engadget adds credibility? Last time I checked they merely copy/paste an excerpt and frame it with their ads. Are we talking about the same engadget?

Also whatever happened to: "Please submit the original source. If a blog post reports on something they found on another site, submit the latter."?

Guess where I quoted that from?

People in glass houses should not throw stones, smokinn.com. However, silly is unimportant. The host/domain primarily just needs to be memorable. Secondarily, it must be entered easily, not offensive, not unrelated to the subject content, and easily found via search. Then it is "good".

Yeah, because Gizmodo and Engadget and TechCrunch and BetaKit and ZDNet and Slashdot all sound so much more professional than Hizook.

My eyes rolled so hard they fell out of my head.

The New York Times original article on Google X placed a lot of emphasis on the lab's work in robotics. I have a feeling Google will be very interested in this challenge. Keep in mind that Google's self driving car project largely grew out of the 2005 Grand Challenge.

http://www.nytimes.com/2011/11/14/technology/at-google-x-a-t...

Yes actually Google hired those that won the 2005 Challenge.

My understanding is that they hired a mix of people who were involved in the challenge, not just the winners.

Part of the Stanley team is there, some of the Red team, and I believe even the guy who attempted to make an autonomous motorcycle is working there.

The mechanism pictured in the linked article -- take that homunculus, fit it with lots of strain gauges, and attach it to a full body haptic feedback harness mounted on a gimbal. Create a mechanism that keeps the pilot's head and body in the same orientation as the homunculus head and body. This enables the pilot's own inner ear to be the stabilization gyros.

Voilà: an animated tethered "Terminator" robot that has eerie human-like balance and motion. The entertainment possibilities for this are awesome. For bonus points, scale it up by a factor of 3 or more. (Anime mecha giant robots, anyone?)

The tech has actually been around for decades.

http://www.raytheon.com/businesses/rids/businesses/gis/strat...

Navigate rough terrain? Drive vehicles? We've been working on this for a while with RoboCup, and walking on felt carpet is challenging. This is exciting, but I hope they don't set the date any sooner than, say, 2018.

No mention of Boston Dynamics in the article?

I think the picture at the top of the article might be from Petman

this... this is not good.

DARPA's increasing funding to their massive tactical decision-making computer (skynet) and only a week later they're putting it to us to build humanoid mechanical soldiers (T-100 series)...

Needless to say, I'm a little worried.

Those images of a humanoid form with arms outstretched are eery especially on Good Friday.

Man, any dutch corps or universities going to participate? I want to work on this!

how about investing LESS money into military research?

Unfortunately, I think LESS money into military research -> LESS money into research altogether. DARPA research is also fairly broad in that its applications tend to span way beyond military-related functionalities.

The D of DARPA is for 'Defense'. It used to be ARPA, but the D was put in front to remind people that it is actually military research.

If you want non-military research, give it to other agencies instead: NSF, ...

Some perspective (that I learned at a "young faculty workshop" recently): DARPA funding is $3B, NSF is $7B, and NIH is $32B. I was surprised.

DARPA is also just one branch of military research agencies.

You're right. I could have been clearer. What I was trying to convey was that the work that goes into these DARPA challenges is still widely used in other areas too.

I'm all for it once you get someone to pay for the research before you spend less on military research. It's really great to want world peace and to be idealistic, but until you can offer real solutions to problems, you're just creating noise by making such comments. The Internet was started by DARPA funding. That's why we're here today.

So, AFTER you've convinced the American tax payer to directly fund this kind of research, or you create an X-Prize crowded funded site, I think we should be thankful that the military is willing to "invest in the future" rather than just build bullets and bombs, which have no positive side effects.

Having a military is a necessary evil in the world in which we live. Find ways to turn some of it into a positive force.

eg. http://www.npr.org/2012/02/17/147047543/desert-military-base...

The military spending has been long gone out of control. Funding of actually useful civilian projects lacks money. Instead DARPA invests into more weapons. We have already enough weapons.

Once again, "military bad" doesn't lead to any sort of useful discussion. Furthermore, complaining about it won't change anything. This is HN and I'd like to think people here solve problems rather they go off on political rants.

Consider that the US economy is $15 trillion. The US handbag market is bigger than DARPA's budget.

http://www.prweb.com/releases/handbags/leather_handbags/prwe...

How can we get more people to invest in science and engineering research? Hopefully, KickStarter and Petridish will eventualy lead to a boom in consumer funded science research.

http://www.petridish.org/

http://www.kickstarter.com/

Hmmmmm, how about quadrupling DARPA's funds? They are miniscule compared to having a navy fleet that's larger than the rest of the world combined.

DARPA is a amazingly great idea, and a lot of their stuff makes their way to civilian applications. More DARPA, less 90,000 ton hardware.

Reply on news.ycombinator.com