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

Protoclone: Bipedal, musculoskeletal Android

clonerobotics.com

Post reply on HN
Print view View original

11–20 of 54 posts

Re: Protoclone: Bipedal, musculoskeletal Android

#11

Post by viraptor »

loanedempathy wrote: post #10
viraptor wrote: post #8

It looks fun, but I'm still amazed by companies trying to create humanoid robots. I mean, any automation, factory improvement, transportation, etc. will be achieved with fewer issues and easier maintenance by purpose built robots. Alternatively by something simple but adaptable (like mounted or mobile arms with different attachments). Unless they're building a sexbot or art... who exactly is the target here? (I get t…

Using humanoid robots means you get to reuse human tools and workflows, and use teleoperation (and training) until your AI catches up.

Why would you want to use human tools? Human tools exist because of how we work. If you're building a robot, it's extremely easier to give it a functionally equivalent attachment instead. (rather than a functionality equivalent arm with enough fine-grained pressure feedback to be able to reliably grab things)

Teleoperation works great on purpose built robots too - for example https://www.bbc.com/future/article/20140516-i-operate-on-peo...

Or phrased a different way, what environment are you thinking of where a fully generalised highly advanced humanoid robot and its maintenance is more cost effective than diying or airshipping a specialised handle?

Re: Protoclone: Bipedal, musculoskeletal Android

#12

Post by quanto »

Their video:

https://www.youtube.com/watch?v=H7dhwFcuUn0

Taken at face value, it is a very cool project. With pneumatic muscles, the biggest challenge is actually not strength but control. Fine strength/distance control is difficult. Pneumatic muscles are more for bang-bang control (on/off). The company seems to address this issue by having many "myofibers" and recruiting the fibers one by one. However, in the video, the movement is still very jerky. My limbic robots were very jerky too.

Nevertheless, very cool. The claimed 500w pump, if really done, is quite a feat. Assuming 80% efficiency, 400w output for muscles would put the robot on par with a competitive cyclist activating quadriceps, the largest muscles in the body.

Re: Protoclone: Bipedal, musculoskeletal Android

#15

Post by HnUser12 »

viraptor wrote: post #8

It looks fun, but I'm still amazed by companies trying to create humanoid robots. I mean, any automation, factory improvement, transportation, etc. will be achieved with fewer issues and easier maintenance by purpose built robots. Alternatively by something simple but adaptable (like mounted or mobile arms with different attachments). Unless they're building a sexbot or art... who exactly is the target here? (I get t…

If the goal is to commoditize, does it not make sense to have a humanoid that you can sell to individual households? Not everyone would want to have a purpose-built robot arm in their kitchen. But this may be an easier sell if it works well.

Re: Protoclone: Bipedal, musculoskeletal Android

#16

Post by viraptor »

HnUser12 wrote: post #15
viraptor wrote: post #8

It looks fun, but I'm still amazed by companies trying to create humanoid robots. I mean, any automation, factory improvement, transportation, etc. will be achieved with fewer issues and easier maintenance by purpose built robots. Alternatively by something simple but adaptable (like mounted or mobile arms with different attachments). Unless they're building a sexbot or art... who exactly is the target here? (I get t…

If the goal is to commoditize, does it not make sense to have a humanoid that you can sell to individual households? Not everyone would want to have a purpose-built robot arm in their kitchen. But this may be an easier sell if it works well.

What's the point of a humanoid in that case? Light frame on rollers would be much safer. There are rolling designs that can deal with stairs too. Humanoids waste lots of energy just balancing themselves and use precise adjustments (toes do quite a bit of work here). And if they fail at balancing, you've got tens to hundreds kilograms of equipment falling through a wall / on your possessions / on somebody. Equipment that may decide to start getting up in that situation and cause more issues...

Re: Protoclone: Bipedal, musculoskeletal Android

#17

Post by thomasfedb »

viraptor wrote: post #11
loanedempathy wrote: post #10

Earlier quoted context omitted.

Using humanoid robots means you get to reuse human tools and workflows, and use teleoperation (and training) until your AI catches up.

Why would you want to use human tools? Human tools exist because of how we work. If you're building a robot, it's extremely easier to give it a functionally equivalent attachment instead. (rather than a functionality equivalent arm with enough fine-grained pressure feedback to be able to reliably grab things) Teleoperation works great on purpose built robots too - for example https://www.bbc.com/future/article/201405…

Probably because we have quite a lot of human tools already.

Re: Protoclone: Bipedal, musculoskeletal Android

#18

Post by ulrikrasmussen »

viraptor wrote: post #11
loanedempathy wrote: post #10

Earlier quoted context omitted.

Using humanoid robots means you get to reuse human tools and workflows, and use teleoperation (and training) until your AI catches up.

Why would you want to use human tools? Human tools exist because of how we work. If you're building a robot, it's extremely easier to give it a functionally equivalent attachment instead. (rather than a functionality equivalent arm with enough fine-grained pressure feedback to be able to reliably grab things) Teleoperation works great on purpose built robots too - for example https://www.bbc.com/future/article/201405…

Assuming the cost of the droid is low enough, I think the use-case is to have it do ad-hoc tasks using the tools you have at hand right now, without the need to invest in an expensive specialized tool which only the robot can use. The idea is not efficiency but versatility.

Re: Protoclone: Bipedal, musculoskeletal Android

#19

Post by jayd16 »

viraptor wrote: post #11
loanedempathy wrote: post #10

Earlier quoted context omitted.

Using humanoid robots means you get to reuse human tools and workflows, and use teleoperation (and training) until your AI catches up.

Why would you want to use human tools? Human tools exist because of how we work. If you're building a robot, it's extremely easier to give it a functionally equivalent attachment instead. (rather than a functionality equivalent arm with enough fine-grained pressure feedback to be able to reliably grab things) Teleoperation works great on purpose built robots too - for example https://www.bbc.com/future/article/201405…

Well I don't know that I quite buy it myself but if you want the thing to work along side humans you can't just put special handles in everything.

If you want a robot to pour you coffee from a coffee maker humans also use, how do you imagine that working?

Re: Protoclone: Bipedal, musculoskeletal Android

#20

Post by Kostic »

viraptor wrote: post #8

It looks fun, but I'm still amazed by companies trying to create humanoid robots. I mean, any automation, factory improvement, transportation, etc. will be achieved with fewer issues and easier maintenance by purpose built robots. Alternatively by something simple but adaptable (like mounted or mobile arms with different attachments). Unless they're building a sexbot or art... who exactly is the target here? (I get t…

If there's potential military use and you don't build it, you risk your adversary doing that.
Post reply on HN