Animal secrets
by Robert Doolan
How does a crab scurry sideways? How does a horse accelerate? And how does a cockroach
run?
The animals that the Creator put on this planet are teaching computer scientists
a few things these days.
When NASA scientists want to explore another planet and collect samples from its
terrain, they need robots that can move freely and pick up variously sized objects.
When surgeons give a patient a bionic hand, it needs to grip well if it is to be
an effective replacement for the lost limb. When the military wants to survey battlefields,
or check riverbeds or oceans for mines, it needs robots that can do the job so human
lives are not lost.
To produce devices that may have to move sideways, roll into a ball, accelerate,
grip or pluck, researchers have increasingly turned to studying the creatures that
roam our earth.
They have used high-speed videos to observe animals in action, computer-connected
surfaces to measure how insects scurry and scramble, and computer simulations to
try to make sense of the data they find.
'We're trying to understand how real critters organize their behaviour', said computer
scientist David Zeltzer from the Massachusetts Institute of Technology.1
One study scrutinized the leg movements of various creatures with two, four, six
or eight legs. Investigators found that all these animals used similar patterns
in locomotion. Two legs of a poodle, three legs of an insect, and four legs of a
crab act as a single unit like one leg of a human.
With regard to speed and stride, it is known that trotting horses speed up when
they change to a gallop. To accelerate, they don't move their legs faster—they
simply take longer strides. But scientists have also found that a mouse and crab
of equal weight will change gaits when each reaches a speed of about one metre per
second, with the same stride frequency. This suggests to researchers that there
are general principles which can be applied to animals with great differences in
body form.
In the late 1980s, scientists at Case Western Reserve University in Cleveland, USA,
began designing an artificial cockroach. For two years they developed it on computer.
It could move its legs independently, open and close its mouth, relay information
it 'sensed' from its antennae, and even detect food.
Then the researchers used the computer information to design actual physical models.
The artificial cockroaches did a reasonable job of imitating the real insects. High-speed
videos showed that a cockroach could run 1.5 metres (almost five feet) per second.
As it increased speed, the cockroach leaned back and ran on only four of its six
legs. To zoom along at top speed, it ran on only its two back legs. The artificial
cockroach was programmed to try to do what the real one did.
All this information has helped enormously in devising robots that work efficiently.
Whether the information is used to explore other planets, to survey hostile environments
on earth, or to make bionic limbs that co-ordinate in the most effective way, the
creatures God has designed have been perfect models.
But computer-designed animals fall far short of the real things. Computer scientist
Randall D. Beer, from Case Western Reserve University, admitted: 'Even the simplest
animals are much more versatile than the most sophisticated artificial intelligence
machines.'2
A cockroach, for instance, flees from a predator when it detects a puff of air.
It works out the wind direction, and acts in less than 60 milliseconds.
Work is continuing on improving the computerized animals. A lot of design and planning
is going into it.
This brings up an observation that many who believe in evolution ignore too readily.
The computer scientists and design engineers are not relying in the slightest on
unplanned, purposeless processes to improve their technology. They are using enormous
intelligence, complex planning and specific goals in their work.
In the same way, it seems unreasonable to expect that unplanned, unproven, purposeless,
trial and error evolutionary processes are an explanation for the marvellous and
intricate design that is apparent in all living creatures.
That's a lesson which many more people would do well to ponder.
References
- Elizabeth Pennisi. 'Robots Go Buggy—Engineers Eye Biology for Better Robot
Designs', Science News, 30 November 1991, p. 361.
- Ibid, p. 363.
|
First published:
