Start from here ' ' '

1
STEALING DESIGN FROM NATURE
Julian Vincent Centre for Biomimetic and Natural
Technologies Department of Mechanical
Engineering The University, BATH,
UK j.f.v.vincent_at_bath.ac.uk
2
Solutions from biology ? ? ? ? ? ? ? ? ? ? ? ? ? ?

• Competition selects and optimises - but for what?
• Optimisations are local - organisms are
  multifunctional, have to work while they grow,
  and are derived from earlier designs
• Optimisation means good enough
• Nature may be solving different problems -
  minimum energy or maximum competitiveness?

3
Start from here . . .

• The abilities of living machines can exceed
  those of man-made ones
• Natures solutions survive
• Physics rules, so we can copy and adapt
• HOW CAN WE TRANSFER THE TECHNOLOGY?

4
(No Transcript)
5
The Lotus effect
6
(No Transcript)
7
Polyimide hairs Scale is 2µm
8
Gecko tape in action Carrying capacity of more
than 200 g/cm2
9
(No Transcript)
10
(No Transcript)
11
A penguin is 40oC inside and -50oC outside
12
A penguin feather
A typical feather
13
(No Transcript)
14
The model feather
15
Pine cone effect
16
Pine cone effect
17
(No Transcript)
18
Heuristics of structural design
19
Insect drills
Modified mouthparts
Egg-laying drills
Lepidoptera A few blood-feeding
moths Diptera Mosquito, midge, gnat, horse-fly
Hemiptera Aphid, cicada, Rhodnius, Triatoma,
Orthoptera Locusts, grasshoppers Hymenoptera Ichn
eumonids, Rhyssa, Sirex
20
Sirex gigas
21
Sirex ovipositor
22
Reciprocating drill
23
Deployable wasp drill
24
(No Transcript)
25
Shock absorption by quills
26
(No Transcript)
27
Holes in tracing paper
Each hole acts as a focus for the dissipation of
(global) strain energy thereby stopping it from
getting to the growing crack tip (local strain
energy sink). Diagrams show areas of damage
around holes.
28
Holes in tracing paper
29
Holes in wood
30
John von Neumann (1903 - 1957)
Inventor of the theoretical Universal Constructor
1950s A Universal Constructor would be a
computer linked to a manufacturing robot. The
combination would be able to copy themselves.
JvN with ENIAC
31
Rapid Prototyping
First patents filed in 1971. There are now many
different technologies.
32
A rapid prototyping machine that will incorporate
electronics
Wood's Metal
33
Replicating Rapid Prototyper
A machine that can make all its own components,
except . . .
34
The RepRap Project

• A three-stage project to make a Universal
  Constructor
• 1. Incorporate electrical conductors,
• 2. Design and make the servo movement axes,
• 3. Design and make the material deposition
  system.
• Then release all the results on the Web under the
  GNU General Public Licence.

35
Evolution
The CAD designs (genotype) have to be available
with the RepRap machine (phenotype) for it to be
able to copy itself.
? People will improve the design. ? Some
improvements will be posted back on the Web under
the GPL. ? Old machines can make new designs. ?
Artificial selection speed, simplicity,
accuracy, fewer added parts... ? 'Speciation',
and runaway symbiotic selection.
36
Economics
It doesn't matter how much the first RepRap
machine costs, all the rest will cost
raw-materials assembly-time. ? Once you
have one, you can have any number. ? No one can
make much money by selling them. ? Material
supply. ? Recycling. ? Bringing manufacturing
capacity to the poorest people.
37
Measuring hierarchical rankleaders and
subordinate behaviour
38
Leaders and subordinate behaviour
39
Normalised Entropy Index as a measure of
collaboration and leadership style
40
Normalised Entropy Index as a measure of group
cohesion
41
4 types of ant management
Activation Solitary ants act
independently Minimum cooperation needed No
exact territory borders, no roads Typical of
ants in a small colony More active ants guide
others
Controlling Teams hierarchically organised
Youngsters led by older experts In space we see
streams of activity, but not built roads. An
early stage of colony development in species with
high social complexity and a large population
(millions)
42
4 types of ant management
Coordination Typical of ants with strict
specialisation Tasks need not be performed
completely As all castes in surplus so
unemployment Hierarchy reduces competition to
allow cooperation
Governing Highest social complexity An old ant,
who has performed many different functions in her
life, stays on the top of the nest cupola and
regulates all functions performed outside the nest
43
Technology transfer

• Can we define the problems more objectively?

44
Teoriya
Theory of
Inventive
Resheniya
Problem
Izobreatatelskikh
Solving
Zadatch
45
The Ideal Result

• The Ideal Result describes the solution to a
  problem, independent of or ignoring the mechanism
  or constraints of the original problem.
• The ideal system occupies no space, has no
  weight, uses nothing..
• The ideal system delivers benefit without cost or
  harm (its very green!)

46
The Power of Ideality

• Start with solutions not problems
• Think Out of the Box
• Remove perceived and real barriers by offering
  wacky alternatives
• Forget inching forward while pondering small
  details - make a great leap

47
IDEAL
Desire . . .
HERE
48
IDEAL
Virtual world
Problem!
Real world
HERE
49
William Blake (1757-1827)
Without contraries . . .
. . . there is no progression
The Marriage of Heaven and Hell (1790-1793)
50
Hegel
The mind wants the truth, but cannot think
without drawing a distinction
Every distinction has two terms, every argument
has a counter-argument
We fix first on the one, then on the other, until
finally we come to rest on the distinction itself.
51
Hegel
This process of alternation and rest is
Dialectical motion, which has three stages
THESIS ANTITHESIS SYNTHESIS
52
Dialectics and conflict
Thesis (affirmation)
Introduce conflict with . .
Antithesis (negation)
Resolve the conflict with . .
Synthesis
53
Thesis - antithesis?
goal 1 - goal 2 target 1 - target 2 goal -
obstacle conflict 1 - conflict 2 function 1 -
function 2 parameter 1 - parameter 2 parameter
improves - parameter degrades life gets better
this way, but gets worse in this direction
54
Someone elses HERE
SYNTHESIS (yields homology)
Someone elses real
Real world
HERE
55
Things
do things
somewhere
56
How do you re-design a leaking flange joint?
Increase bolt tightening torque?
Increase the number of bolts?
Introduce/modify gasket?
57
TYPICAL FLANGE JOINT DESIGN TRADE-OFFS
high
bad
Number of Bolts
?
Substance
good
low
good
bad
good
bad
Substance
Leakage Performance
bad
high
?
Bolt Torque
Energy
good
low
good
bad
good
bad
Substance
Leakage Performance
58
The SYNTHESIS box
Degrades gtgt
Improves gtgt
A source of homologues
59
Technological models
60
Technological models
6 26 27 40
12 18 19
10 31
61
Biological models
62
Biological models
13 15 17 20 40
3 6 9 25 35
31
63
Compare tech- and bio-

• 10 - Preliminary action

Overlap between tech- and bio-

• 31 - porous materials (space)

64
(No Transcript)
65
(No Transcript)
66
So . . . What to do about it?
67
Biology Technology

• Dry, rigid
• Homogeneous
• Isotropic
• Rectilinear
• Metallic
• Factory gtgtgt product
• Limited functionality
• Repair or replace

• Wet, flexible
• Heterogeneous
• Anisotropic
• Curved
• Non-metallic
• Factory ltltlt product
• Multifunctional
• Self-repairing

68
Lessons

• Its possible to learn from nature
• Huge changes in context are possible
• Most of natures design can be (carefully!)
  dumped
• Biologists are essential to differentiate
  functions
• A virtuous circle exists between bio- and tech-
• Bio-solutions have control built in to the
  material and the design

69
Successful biomimetics

• Biologist required who must be able to . . .
• . . . identify essential functions
• . . . recognise evolutionary baggage
• . . . recognise developmental baggage
• . . . recognise metabolic baggage
• . . . talk to non-biologists

70
Recommendations

• True interdisciplinary team needed
• The biologist must be there at all times
• Expect unexpected solutions
• Recognise that many solutions are not used by
  nature . . .
• . . . and that natural solutions may be used
  non-optimally
• Frame problems as FUNCTIONS