Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

MIT cheetah robot lands the running jump [video]

youtube.com

Post reply on HN
Print view View original

161–170 of 212 posts

Re: MIT cheetah robot lands the running jump [video]

#161

Post by maxerickson »

IgorPartola wrote: post #140
anthony_d wrote: post #135

Earlier quoted context omitted.

I'm not sure about that. If your car had uncontrolled acceleration, for instance, would you consider use a crowd of people to slow you down or would you put it in the ditch? For the sake of argument that amounts to a left or right turn.

Say there are 10 people in this crowd. The optimal solution for the 11 of us is obviously to have my car drive off a cliff to the right, not into the crowd on the left. However, for me, the optimal solution is obviously to not drive off a cliff. However, your argument omits a couple of interesting details. First off, if a car can have uncontrolled acceleration, it can likely have any number of glitches. It may for ex…

Humans typically avoid creating the choice between crowds to drive into (by following traffic laws, slowing down when there are obstructions in the street, etc.).

I haven't tried to figure out what the numbers are, but my gut expectation is that vehicles driving into crowds is so rare that there aren't really statistics about it.

Re: MIT cheetah robot lands the running jump [video]

#162

Post by Animats »

vlasev wrote: post #31

No one seems to be discussing how this can be improved! Here are a few points: 1. Real cheetahs don't run like that. The action of both pairs of legs is staggered for smoother motion and control. Check this out - https://www.youtube.com/watch?v=131wvVGjZUc 2. Looking at this jumping-over-obstacles video, it seems like the robot is lacking flexibility. The real life cheetah's hind legs go quite a bit underneath in pre…

That's not Boston Dynamics' Cheetah. This is.[1] This new video is MIT's cheetah. It's good to see MIT doing serious legged locomotion work again. Things stalled out there after Raibert left and Gil Pratt took over. Looks like DARPA is funding MIT, now that Google bought Boston Dynamics and stopped taking DoD contracts.

Not sure why they chose a pronk gait, except that they're doing all straight line work. A gallop has more turn options, but roll control is easier for the pronk. Hopping over an obstacle has been done before, by some of the early MIT planar bipeds.[2]

It took about $125 million in DoD funding to get a usable Big Dog and Atlas. It's quite possible to do a lot more in legged robotics, but it is not yet cheap. It's still not clear if Google will achieve commercial legged robots, or decide to "put more wood behind fewer arrows" again and focus on their core business, advertising. The commercial payoff is a long way ahead.

Academic projects rarely have the funding and staff to get beyond the demo level. You need some big assets, like machine shops and skilled machinists. Note that they're borrowing MIT's gym. To get real work done, you need full time access to your own test area. It's hard to get that on a college campus. The typical academic robotics project is one professor and three grad students.

[1] https://www.youtube.com/watch?v=chPanW0QWhA [2] https://www.youtube.com/watch?v=XFXj81mvInc

Re: MIT cheetah robot lands the running jump [video]

#163

Post by tunesmith »

deepnet wrote: post #88

Stuart Russel of AI a Modern Approach fame ( with Norvig ) asks us to consider the ethics of AI research - killer robotics, & hunter drones. http://www.nature.com/news/robotics-ethics-of-artificial-int... He calls for a moritorium until we can get a neural net to learn Asimov's 1st Law. Should the self driving car swerve into you to avoid greater moral hazard ? Humour aside as researchers and robotocists do we have a…

I wish I had a source, but I seem to remember an engineer for google's self-driving cars or a precursor of them saying that in those cases, the only solution they could come up was to switch off the car's decision making and let nature take its course. In the Trolley Problem, this is akin to letting the train mow down the three people so the decision maker doesn't have to take the accountability for choosing to kill the one other person. Which is a pretty common answer. (Most people wouldn't push a person off a bridge to block the trolley, either.)

Re: MIT cheetah robot lands the running jump [video]

#164

Post by biot »

chinathrow wrote: post #49
DanBC wrote: post #37

Earlier quoted context omitted.

Wait what? Those expensive machines make war harder, not easier. Central African Republic has war with child soldiers using machetes and guns. It's an unpleasant truth: humans are disposable. This is evidenced by a bunch of stuff - rich westerners have their clothes made by poor people locked into unsafe Bangladeshi warehouses; have many of our goods made by low people in terrible and dangerous conditions; by our lac…

Those expensive machines makes not a single child soldier go away - why should it?

Your hypothesis was that machines lower the entry level for war. The counterexample of using untrained, disposable child soldiers indicates that the entry level is already zero. Consequently, your hypothesis is wrong; the entry level for war is actually raised via expensive machines.

Re: MIT cheetah robot lands the running jump [video]

#166

Post by lotu »

IgorPartola wrote: post #120
deepnet wrote: post #88

Stuart Russel of AI a Modern Approach fame ( with Norvig ) asks us to consider the ethics of AI research - killer robotics, & hunter drones. http://www.nature.com/news/robotics-ethics-of-artificial-int... He calls for a moritorium until we can get a neural net to learn Asimov's 1st Law. Should the self driving car swerve into you to avoid greater moral hazard ? Humour aside as researchers and robotocists do we have a…

> Should the self driving car swerve into you to avoid greater moral hazard? The self-driving car's first priority should be to protect those inside it. Anything else is more complicated. I read an article a little while ago about this topic and it posed a question along the lines of "should a self-driving car do the thing that'll kill its own passengers, yet save a larger number of others around it?" Think about how…

Realistically the self-driving car isn't going to be smart or fast enough to make these choices. We are assuming some infallible perfect AI when if-fact you have a bunch of (relatively) simple algorithms a bunch of mapping data. The fact is that the type of complicated moral analysis will take time prehaps in the absolute best case several 1/10th of a second. For a moving vehicle that analysis will likely be moot during that time. We don't expect human drivers to make such difficulty snap decisions as part of a driving tests why should we ask a self driving car to?

Re: MIT cheetah robot lands the running jump [video]

#167

Post by jameshart »

kbenson wrote: post #126
posnet wrote: post #24

Earlier quoted context omitted.

Which leads to the appeal of a mechanized infantry that never gets tired, never gets sick, kills without remorse, follows orders invariantly, doesn't eat or drink, has an upgrade path but most importantly can't be killed only destroyed or broken. Sure the first generation won't replace all soldiers and they may never replace them all. We will need logistics and support people of course. Its not a perfect solution. Bu…

I'm all for this, not just for our soldiers, but also for the remote population. Part of the problem of actual people in these situations is that they fear for their lives, so make decisions based on their own fear and keeping themselves alive. Eventually, robotic soldiers can hopefully be better than human soldiers at identifying a threat, even if it's just because they are programmed to take more time to do so. The…

Right - a robot might kill without remorse, but it also never kills because of fear, or hate, or to take revenge for its fallen comrades.

But then why does it kill at all? Presumably because it's been sent to further a human agenda.

If you imagine a noble military purpose - liberating an oppressed population from an aggressive occupying enemy, for example - then your robot soldiers are awesome; they will target only combatants; minimize collateral civilian casualties; they will never loot or rape; they will selflessly interpose themselves between the innocent and those who would harm them. The perfect heroic soldier, better than any human army could be. They will be greeted as liberators.

But to the extent that the underlying human agenda involves pacifying a civilian population, instilling fear, or outright causing terror, there's no reason to think that a robot soldier would not be capable of being far worse than human soldiers. It can't be reasoned with. It doesn't have a conscience. It doesn't matter that it doesn't 'fear for its life' when kids throw stones at it if it's been programmed to respond to that threat with deadly force precisely to discourage other kids from throwing stones.

Re: MIT cheetah robot lands the running jump [video]

#168

Post by IgorPartola »

malkia wrote: post #149
IgorPartola wrote: post #120

Earlier quoted context omitted.

> Should the self driving car swerve into you to avoid greater moral hazard? The self-driving car's first priority should be to protect those inside it. Anything else is more complicated. I read an article a little while ago about this topic and it posed a question along the lines of "should a self-driving car do the thing that'll kill its own passengers, yet save a larger number of others around it?" Think about how…

Most-obvious, but a self-driving would be riding "alone" for long periods of time. For example "sending" car from one member of the family to another (or one client to another for use), or maybe delivering items from the store directly into a box. So in this case, the car should have the other mode, which is not safe for the car, but from whatever is outside, but now that I've typed it, even that I don't know the det…

Ooh, I really like this idea. Instead of me finding parking at the airport, I send my car to park in my own driveway, then have it pick me up a few days later. That's brilliant. I haven't even thought of the possibility of having the car drive empty.

Re: MIT cheetah robot lands the running jump [video]

#169

Post by hyperion2010 »

deepnet wrote: post #88

Stuart Russel of AI a Modern Approach fame ( with Norvig ) asks us to consider the ethics of AI research - killer robotics, & hunter drones. http://www.nature.com/news/robotics-ethics-of-artificial-int... He calls for a moritorium until we can get a neural net to learn Asimov's 1st Law. Should the self driving car swerve into you to avoid greater moral hazard ? Humour aside as researchers and robotocists do we have a…

Implementation of the 0th law requires the implementation of a utilitarian calculus. Also, shouldn't we put a moratorium on having children until humans can learn how to not kill each other?

edit: if you're going to downvote please respond

Re: MIT cheetah robot lands the running jump [video]

#170

Post by JoeAltmaier »

deepnet wrote: post #88

Stuart Russel of AI a Modern Approach fame ( with Norvig ) asks us to consider the ethics of AI research - killer robotics, & hunter drones. http://www.nature.com/news/robotics-ethics-of-artificial-int... He calls for a moritorium until we can get a neural net to learn Asimov's 1st Law. Should the self driving car swerve into you to avoid greater moral hazard ? Humour aside as researchers and robotocists do we have a…

Its locking the barn door after the cow is gone. 100,000,000 killer robots are already deployed, killing an average of 4000 innocents each year. Land mines, which kill indiscriminately. We've made negligible progress controlling their deployment.
Post reply on HN