Welcome! Log In Create A New Profile

Advanced

Better self-replicating through 'flexonics'

Posted by SebastienBailard 
Better self-replicating through 'flexonics'
December 17, 2007 03:01AM
This is the current paper I'm reading:
www.cs.berkeley.edu/~risnerj/flexonics/wafr2002.pdf
A bit more at:
www.cs.berkeley.edu/~risnerj/flexonics/

Instead of using traditional motors, joints and sliding parts, and such, the idea is to print everything at once, using electroactive polymers and strategic flexixion (bending). The researchers have ended up building an inkjet-based 3D printer because they found fdm too limiting for the scale they want to work at.

This is of interest, because it suggests we could print RepRaps with printed motors and joints, and thus be more perfectly self-replicating. This would be for Mendel, not Darwin (the 1.0 release).
VDX
Re: Better self-replicating through 'flexonics'
December 17, 2007 05:05AM
Hi Sebastien,

... this is the main point in my interest in parallelkinematic systems and the discussions in the threads "Artificial Muscles" and other related too ...

I already developed and built some flexure-structures for micro-grippers and nanopositioners with piezo-actuators and hydraulic or magnetic driving.

With some other micro- and nano-mechanic setups i can drive and move makroscopic systems with a simple oszillating pump or solenoid, so with a bit more 'evolving' we haven't to handle with complex and hard to reprap motors or other heavily sophisticated stuff - simply design 'soft' and/or bionic-shaped objects and print them winking smiley

Viktor
VDX
Re: Better self-replicating through 'flexonics'
December 18, 2007 02:39AM
... i emailed John Canny (the Prof. behind flexonics) if he have some more actual infos or is interested in participating ...

'Soft' mechanical systems in comparison to 'rigid' ones are more interesting when you go down in the micro- and nano-scale, as you can't reduce the sizes and accuracies of bearings and other mechanic and electromechanic elements under a specific limit - either the costs rise exponentially, or physical properties (friction, adhesion, inductance ...) changes drastically and some 'normal' mesoscopic function didn't work anymore at micro-scale.

So when the building-accuracy of the reprap undergoes the 0,1mm drastically (i'm actually in the 0,01mm-range and try to rise the accuracy even more), flexonics or bionics would be adressed directly ...

Viktor
Re: Better self-replicating through 'flexonics'
December 18, 2007 08:41PM
Email Jeremy Risner too. As a grad student, he may have the free time to compose more than a polite 3 line response.

It would be extremely useful to bring in some folk with experience, and they just built such a machine.
VDX
Re: Better self-replicating through 'flexonics'
December 19, 2007 01:51AM
... OK, triggered ...

Viktor
Sorry, only registered users may post in this forum.

Click here to login