The ghost is the machine - simple cybernetics making things move
June 19, 2020 8:23 PM Subscribe
While engineering gets more complicated, more opaque and less complex/redundant (Complicated or complex – the difference is important), there's a strand of making where IT is used in system design, but not centrally present in the finished item a softer side of science - an engineering ideas rabbit hole.
Autonomous systems that respond to where they are, through deep exploration of materials and potential inputs, and then built in at the molecular/material level.
I've avoided a 'metamaterial' focus as this is often more about meta and less about material; I want to explore materials with real applications and hopefully some actual physical objects
This fast-changing field is branching so fast I had to set some rules to stop my FPP wandering off:
- eyeball visible changes
- fully autonomous
- very low CPU use - ideally sensors as a property of the material
- no black box / tells its own story
- mainly homogeneous / fabric / natural materials based. Preferably no mechanicals
- function embodied through deep system knowledge
Abdelmohsen, 2016 puts it well: "soft (less dependent on hard mechanical systems) and responsive (dynamically and simultaneously adapting to spatial and environmental conditions)".
Approaches here are as robust and simple as possible as multipart linkages tend to break, so these are mainly monolithic, or sheet/fabric systems, for instance
Shape morphing Kirigami mechanical metamaterials.
The Grumman X-29 was an early example of very robust so-called aero-elastic tailoring where a flexible wing component changed shape with stresses with only air flow as an input.
Self-crumpling/uncrumpling magnetic fabrics offer ways to catch, grasp and fold around objects as well as explosively crumpling, offering spring-like activity.
Weather-responsive especially timber building parts e.g. louvres, shades, ducting, tracking...
https://honr.ethz.ch/en/the-group/facade/wooden-bilayer.html wooden solar-PV tracking system mefi couldn't handle link - use 'Subnavigation' button for more soft/natural materials robotics.
Every time I look at this topic a new branch has been started so I hope to write more FPPs on this area. Others areas include stormwater, acoustic control, earthquake control, beach-sand stabilisation...
Autonomous systems that respond to where they are, through deep exploration of materials and potential inputs, and then built in at the molecular/material level.
I've avoided a 'metamaterial' focus as this is often more about meta and less about material; I want to explore materials with real applications and hopefully some actual physical objects
This fast-changing field is branching so fast I had to set some rules to stop my FPP wandering off:
- eyeball visible changes
- fully autonomous
- very low CPU use - ideally sensors as a property of the material
- no black box / tells its own story
- mainly homogeneous / fabric / natural materials based. Preferably no mechanicals
- function embodied through deep system knowledge
Abdelmohsen, 2016 puts it well: "soft (less dependent on hard mechanical systems) and responsive (dynamically and simultaneously adapting to spatial and environmental conditions)".
Approaches here are as robust and simple as possible as multipart linkages tend to break, so these are mainly monolithic, or sheet/fabric systems, for instance
Shape morphing Kirigami mechanical metamaterials.
The Grumman X-29 was an early example of very robust so-called aero-elastic tailoring where a flexible wing component changed shape with stresses with only air flow as an input.
Self-crumpling/uncrumpling magnetic fabrics offer ways to catch, grasp and fold around objects as well as explosively crumpling, offering spring-like activity.
Weather-responsive especially timber building parts e.g. louvres, shades, ducting, tracking...
https://honr.ethz.ch/en/the-group/facade/wooden-bilayer.html wooden solar-PV tracking system mefi couldn't handle link - use 'Subnavigation' button for more soft/natural materials robotics.
Every time I look at this topic a new branch has been started so I hope to write more FPPs on this area. Others areas include stormwater, acoustic control, earthquake control, beach-sand stabilisation...
From the Kirigami mechanicals link: "... the shape-shifting of thin, planar sheets (which we consider 2D) into 3D structures ..."
This is a technique with a long and robust pedigree. Life has been based on essentially 1D molecular chains that automatically shape-shift into 3D structures since before it was life.
posted by flabdablet at 12:54 AM on June 20, 2020 [1 favorite]
This is a technique with a long and robust pedigree. Life has been based on essentially 1D molecular chains that automatically shape-shift into 3D structures since before it was life.
posted by flabdablet at 12:54 AM on June 20, 2020 [1 favorite]
built in at the molecular/material level
i wonder where the line is where it crosses into bioengineering?
posted by kliuless at 3:59 AM on June 20, 2020 [1 favorite]
i wonder where the line is where it crosses into bioengineering?
posted by kliuless at 3:59 AM on June 20, 2020 [1 favorite]
I had some problem with some of the writing, the example of making bread vs moon shot, vs raising a kid seem unwellconsidered. But love love love one article title:
Imagine you are a Sea Slug Larva…
posted by sammyo at 4:54 AM on June 20, 2020 [1 favorite]
Imagine you are a Sea Slug Larva…
posted by sammyo at 4:54 AM on June 20, 2020 [1 favorite]
Really enjoying the rabbit hole on systemic design and the rest on Learning for Sustainability too, thanks for the link!
“We can determine complicated outcomes. We can only enable complex outcomes. We can specify complicated systems. We can only intervene in complex systems.”
posted by brism at 5:47 AM on June 20, 2020 [3 favorites]
“We can determine complicated outcomes. We can only enable complex outcomes. We can specify complicated systems. We can only intervene in complex systems.”
posted by brism at 5:47 AM on June 20, 2020 [3 favorites]
I would pay cash money for the opportunity to vote for somebody with the wit to understand simple/complicated/complex, the charisma to explain it to people without the wit, and a stockpile of knowledge of skeletons in closets of those who would sabotage effective campaigns to influence things to the benefit of the poor, disenfranchised, and other 99%ers.
posted by spacewrench at 8:48 AM on June 20, 2020 [4 favorites]
posted by spacewrench at 8:48 AM on June 20, 2020 [4 favorites]
sixswitch - thanks, I hadn't found that, both mesmerizing and creepy.
kliuless - yes, well it's very hard to know really; when you find ways to dispense with wheels and materials/systems are flexible and monolithic and without linkages it all becomes very blurred.
brism - yes it's a good page. Some of the messier things I get to do involve shifting things from a non-redundant state to a linked/networked state. It seems we are all taught from the earliest age that redundancy is a bad thing, when it is essential
spacewrench - Well I've got the first bit, sadly I lack charisma, and have been told I'm humourless.
posted by unearthed at 3:08 PM on July 1, 2020
kliuless - yes, well it's very hard to know really; when you find ways to dispense with wheels and materials/systems are flexible and monolithic and without linkages it all becomes very blurred.
brism - yes it's a good page. Some of the messier things I get to do involve shifting things from a non-redundant state to a linked/networked state. It seems we are all taught from the earliest age that redundancy is a bad thing, when it is essential
spacewrench - Well I've got the first bit, sadly I lack charisma, and have been told I'm humourless.
posted by unearthed at 3:08 PM on July 1, 2020
« Older How to make Garum, the most famous of ancient... | Peter did not answer. Because, after all, it was a... Newer »
This thread has been archived and is closed to new comments
Check out this lil’ hexapod!
posted by sixswitch at 9:40 PM on June 19, 2020