DARwIn-OP unveiling!!!

[Gort]

DARwIn-OP unveiling at the The 2010 IEEE-RAS International Conference on Humanoid Robots.

<table><tr><td></td></tr><tr><td>From I, Bioloid</td></tr></table>

http://mike-ibioloid.blogspot.com/2010/ ... l?spref=fb

[SK]

That looks very nice, good to see some competition for Nao
I'll be at Humanoids 2010 too, will attend another workshop though.

[SK]

Larger photo here:
http://darwinop.sourceforge.net/

and the sourceforge presence that's in the process of being filled with content:
http://sourceforge.net/projects/darwinop/

[JavaRN]

Hope it's affordable somehow more than NAO is. From what I saw in the photo the servos used are not AX18F like in the original model but RX28 or something of the kind.

[Gort]

Yes, that looks like the new Version of the DARwIn-OP!



http://mike-ibioloid.blogspot.com/2010/ ... in-op.html

[siempre.aprendiendo]

Woow it looks great! The new version from the same source:




This video is somewhat funny

phpBB [media]

[billyzelsnack]

This spec says that it uses the RX-28M.

http://sourceforge.net/projects/darwino ... f/download

RX-28:
16V - 37.7kgf-cm @ 79rpm
12V - 28.3kgf-cm @ 60rpm
RS-485 interface

RX-28M:
12V - 24kgf-cm @ 45rpm
TTL interface up to 4.5MBPS

If you normalized by speed it would put the torque at 18kgf-cm. So it seems like it is more like an overweight AX-12+ in a RX-28 package. Unless the RX-28M can do 16V and RS485 it does not seem like it really is in the same league as the RX-28.

[Gort]

Found hardware and software specs on the DARwIn-OP support website!!!!!!

New servos, two controllers, 6-Axis IMU and even more check it out!!!!


http://mike-ibioloid.blogspot.com/2010/ ... about.html

[i-Bot]

I was disapointed to see the sourceforge project full of glossy marketing blurb, rather that what we normally see in an opensource robotics project. It always appears that educational establishments are worse at sharing than individuals.

Like Nao, I am not convinced we are going to see much technology breakthrough here, but this is an exciting phase in the attempt to commercialise small humanoids.

For the servo, it looks like the M suffix is for the magnetic potentiometer. Also with a 4.5Mbps data rate, is this a change in servo processor too ?

[UncleBob]

The hardware is basically a netbook pc which has huge amount of processing power and memory. I am more interest in how it can be put into use though. Ie how the processing power can be use for image processing.

It would be a great step with robotis would develop a simple to use software library for image processing which a simple hobbist like me can program with and so advance stuff like picking up a bottle

[SK]

I was disapointed to see the sourceforge project full of glossy marketing blurb, rather that what we normally see in an opensource robotics project. It always appears that educational establishments are worse at sharing than individuals.

I'd tend to cut them some slack. They're just starting that stuff, the official presentation will be in 2 weeks in Nashville, so I don't think one should jump to conclusions just yet.

Like Nao, I am not convinced we are going to see much technology breakthrough here, but this is an exciting phase in the attempt to commercialise small humanoids.

Both Nao and DARwIn-OP certainly don't represent technological breakthroughs, cutting edge technology in humanoid locomotion research generally tends to be expensive and not very robust. Instead, they provide largely proven technology in a complete package, which is worth a lot for people who want to concentrate on different stuff than hardware design and the accompanying hassles.
From my (autonomous robot soccer) POV, DARwIn-OP looks more promising than Nao as it has adequate processing power from the get go, as well as one good camera.

[roboticage]

NAO has it's limitations but at least it exists and has a much more bigger community to support and development at the moment!
It is better commercial looking also ... didn't you think that design it's part of technology also? Why QRIO and ASIMO can move and look good at the same time? It's to hard to make an good looking walking WiFi-Gyro-Accelerometer platform? It's ROBONOVA the final design?
I think competition will make robotics as any other field of activity better and better year after year but it's premature to discus and even more to compare it to NAO!
... personal opinion ... no hart feelings ...

P.S. For the record NAO do not cost double ( ??? 16000 EUURO) comparing with DARwIn-OP (8000 EURO) ... if interested few "jobs" are open on the NAO Developer Program for only 3600 EURO! ... hurry up...

Horia Pernea
http://www.roboticage.eu

[i-Bot]

That Nao is "good looking" and "better commercial looking" is for the customers to decide. I am dissapointed that both the Nao and Darwin have chosen the same dorky childish style. They both appear to portray a submissive doll like persona. Their shuffling movements are aligned with this and equally very unattractive. Good design, maybe, breakthrough, Non.

Asimo at least shows more character and attitude. Many of the smaller Japanese robots also are starting to be more adventurous in different design styles, and being more innovative in their impact on human interactions.

At least we are past the Furby stage.

[Gort]

I will be in Nashville for the unveiling of the DARwIn-OP. I will try to find out as much as I can and take many pictures and videos of the robot in action!



http://mike-ibioloid.blogspot.com/2010/ ... -ieee.html

[SK]

It is better commercial looking also ... didn't you think that design it's part of technology also?

That of course largely depends on what you want to do with the platform. My personal preference is on performance (processing power, locomotion, camera image quality..) over looks, at least when both are competing and you have to trade-off one for the other.
I have the feeling that in some respects, the opposite approach was taken for Nao, with a heavy emphasis on appearance. This of course is great for fields like Human-Robot-Interaction, but not necessarily for all applications.