Ladies and Gentlemen:
I would like to
thank Perry Blackshear for his lyrical prose in putting together his announcement
of today's meeting. He should obviously have at least equal time today,
or in a later engagement to introduce and defend his list of bioengineering
heroes, several of them unknown to me.Today
I would like to confine myself to my original title - simply Biomimetic
Engineering. The concept is simple - Biomimetic Engineering is Engineering
done deliberately in the image of life - Bio from the (Greek) root Bios
meaning life, and mimetic from the Greek mimos and the Latin mimus, in imitation
of, as in mimeograph, mime, mimic, etc. I perceive this as a major issue,
not a small semantic argument. Indeed, I perceive much of our modern engineering
derived from Physics, Chemistry, and their special brands of Mathematics
as laundered versions of Biomimetic Engineering, laundered in the sense
that questionable money is traded, sometimes at a discount, for unquestionably
clean dollars coming from authoritative and proper sources. I would like
to raise the credibility and the scientific credit rating properly algorithmic
science and technology arising out of biology, medicine, agriculture and
even the behavioral and social sciences to where they could legitimately
create new concepts, new computers and mathematical "figures of thought"
without cumbersome and inexact translation into standard hard science dialects.Before
we go into the nitty gritty details of the proposed matrix inversion, let
me stop to ask a fundamental question, "What is engineering?"
Do you all have an instant utility answer? I was really surprised when a
I asked several groups of engineering students and even assorted staff members
this question, to find that the central idea of applying science and technology
to the solution of human problems and the serving of human problems, not
immediately evident. Let me offer you a typical dictionary definition (l).
This definition obviously excludes the concept of the locomotive engineer
and the person who contrives to make something happen as in "engineering
a public confrontation between two antagonists".I
cannot resist the temptation of showing you some of the youthful definitions
of an engineer that a recent MIT study revealed. (2)Joking
aside, this recognition that Engineering is what Engineers do, not what
the dictionary says that engineers do, has a real bearing on my message
today. I hope to address four major issues. First there is the question
of reestablishing third level education in our American higher education
systems. Next there is the examination of our present position in the industrial
revolution sequence. Third there is the topic of biomimetic invention and
innovation and their management, and finally there is the bottom line question,
"Can we and should we reinvent our Scientific - technical way of thinking
from the technical discipline categorization back to the problem oriented
form analogous to the problem oriented medical way of thinking that is making
considerable headway today?"Let
me elaborate slightly on my rather cryptic comment on restoring third level
education to our higher educational curriculum.  I
referred to a public lecture that I gave a couple of years ago in which
I attempted to separate out four chronobiologically sequential phases of
education. These levels seem related to biological "windows" of
aptitude and susceptibility to learning that open and close at roughly specifiable
times during the development of an individual.We
all know the extraordinary ability of children to learn a complex human
language such as English or Chinese or even Finnish. Indeed all three might
be learned simultaneously without special tutoring through simple social
contact with trial and error. A few years later this would be nearly impossible
and would require intensive effort and a strong accent would almost surely
remain. My post-doctoral Japanese students have uniformly avoided trying
to teach me Japanese while they invited me to improve their early-school-learned
English.K. Lorenz,
in his 1970 Harvard Press book "Studies in Human and Animal Behavior",
retells the story of his phenomenally successful experiments in imprinting
birds with lasting behavioral patterns by introducing key signals at optimal
"window" phases of development.I
observed that we almost intuitively teach children at level one the rudiments
of who is Mama, who is a policeman, how to speak, to read and write, and
how to use a bicycle or a bathroom. I was most amused recently in Japan
to find a small diagrammatic plaque beside the roll of tissue in a western
style bathroom teaching how to use a western style toilet. I also observed
in an advanced course teaching physicians to read diagnostic electrocardiographic
traces, that the students learned the art successfully from a good experienced
teacher in spite of a step by step instructional manual which they learned
to evade in technical detail where it would have led to errors.Second
level education, in my book then, is education in "how to do it"
at a higher level. How
to be a lawyer, how to be a control systems electrical engineer, how to
get good grades or a high salary without working too hard. These educational
processes are necessary and desirable, and I believe we do a good job in
meeting the needs in undergraduate college and in post-graduate education.Level
three education, which includes invention, innovation, and discovery, I
believe we have let slip, in our effort to automate education prematurely
before we know the biomimetic principles of apprenticeship learning. While
I applaud the introduction in a few universities of courses on introduction
to research, to innovation, to invention, I am convinced that much of the
actual high level learning in this domain is of the apprenticeship type
including learning by example even with a faulty rationale to justify the
steps.History
is full of this learning-to-create-by-apprenticeship. Socrates (470-399
BC) -' Plato from age 20 (427-397 BC) + Aristotle (384-322 BC), H. Davy
(1778-1829) -~ his assistant Michael Faraday (1791-1867), Tyndal]. (1820-1893).
Apparently the time slot between the mid-teens and the early twenties is
the one where inventiveness and creativity can be easily learned and, like
bicycle riding, never forgot. I personally had the opportunity in this age
slot to work under the direction of several of these near genius innovators
and I will assure you it is a learning experience never to be forgot.Now
I want to elaborate on the several components of third level education,
emphasizing in particular the biomimetic aspects of each track, but I would
like to take a moment to define level four education so that that will not
be ) left as a cliff-hanger.Level
four education I do not know how to implement. It is a rare but wonderful
process whereby a few remarkable multifaceted individuals put together a
myriad of differently expressed truths into a universal generality. Newton
was this kind of individual who created mechanics, optics, mathematics,
and a rational gravitational theory. Aristotle was an ancient exponent of
this skill, Einstein a recent one. Notice, however, that each of these giants
has left a track so sacred that it is only with temerity that we undertake
to update the masters. Biomimetic
updates are in order, by the way, and I would be glad to point out these
[unreadable]. Returning
to level three; consider the proposition that Invention and Innovation can
be effectively classified within three categories: 1) devices, 2) systems,
and 3) algorithms. We can even discover several different strategies by
which each of these classes of invention and innovation can be carried out
and can identify which one or ones are in the repertoire of a specific practitioner
of the art (or science). Because devices are the central focus of invention
in the public concept and even in the thinking of bioengineers and engineers
in general, the other two categories suffer through lack of recognition
even though all three are comparable in extent and importance.To
make clear the complementary but distinct, roles of invention and innovation,
just think of the many discoveries, feasibility proven designs and even
patented devices of real merit that have never reached the market place,
having failed in the innovation processes. Many of us who routinely and
skillfully invent and bioengineer devices and systems for clinical and non-clinical
medical purposes, are remarkably oblivious to the large role that marketing,
scientific and educational as well as commercial, must play in innovative
introduction of a biomedical device or system.This
role of marketing in the rapid advance of a technology and acceptance of
its potential benefits by our involved public, a government, an Industry
and academia, is clearly understood by some of our international counterparts.
I recently had the opportunity to be a member of an American trade mission
to Japan to examine the technology, the management, and the government relationships
with Industry and with universities in the area ofRobotics,
and it was very evident that the tripartite non-adversary systems policy
linking government, industry and academia was working well and was very
productive. Similar,
but less systematically unified, programs are being established successfully
in West Germany and in several other European countries. Had we the time
I would like to show you the [unreadable] for a center for innovation and
technology utilization that we could [unreadable].We
suffer from an inherited, originally good system of checks and balances
that is intrinsically inhibitory because of its adversary basis. We have
not learned the opposite strategy of mutually supportive contrivance for
success. We have open to us, if we will only seize the opportunity, a third
stage of progress for which we must do systems planning in medical as in
other domestic areas. This is the strategy combining the merits of the old
checks and balances philosophy with the new mutually supportive non-adversary
designs. The answer is multi-beneficiary co-optimization supported by specially
developed quantitative algorithms at a good engineering level of approximate
quantization.Marketing
of a new system design of any consequence in the biomedical fields is by
its very nature a slow process for which we do not at present have a regular
national mechanism, although our National Academies could potentially serve
in this capacity. A concept must be heard, heard again and yet again, in
ever new configurations to avoid staleness, until eventually the hearer
reinvents the Idea for himself. A typical time for emergence is 13 - 15
years. As this exceeds the time allowed for any ordinary researcher, engineer
or governmental administrator to show results, a separate mechanism should
be established to handle this need. I recall vividly an early attempt, about
a decade ago, to introduce this system optimization concept through the
medium of an Engineering Foundation Conference. At the end of the week,
with roughly a hundred participants, most of them highly qualified in bioengineering,
medical and related fields, only two or three could accept the plausibility
of a designed quantitative co-opimal health care delivery system. The plausibility
is rising and we have much of the technology and some of the algorithmic
invention needed, so that the time is ripe for biomedical system innovation.
The Santosha Index may yet be heard throughout the land!I
noted with interest that a surprisingly large part of the Tokyo International
Robotics Congress and exhibition last October was directed toward Medical
Robotics. While I personally thought that some if the designs such as the
robotic seeing eye dog to lead the blind and the patient handling robot
to serve paraplegic and other handicapped patients were premature technically,
there was clearly present the conviction that tasks such as these were about
to be undertaken seriously and successfully in the near future. This is
the attitude that is building automobiles robotically so successfully. I
shall try to offer you specific biomimetic examples from personal experience
in each of the categories at level three but before that I must locate us
technologically in the successive "windows" of biomimetic opportunity.You
will probably remember the rather profound biological principle of Ontological
recapitulation of Phylogreny. Loosely restated, higher organisms, during
their embryological, neonate and youthful periods of development, progressively
go through anatomical, control and learning processes successfully in stages
closely resembling progressively more elaborately organized life forms.We
can see an amazing similarity to this process in the cultural technology
through which we are evolving. Being immersed in it, we do not perceive
the change, just as we, as children, never realized that we had innate abilities
that were not being developed because the imprinting didn't happen. Probably
all of us have or had, remarkable [unreadable] abilities commonly regarded
as genius, psychic or superconscious which we were culturally inhibited
from developing.Returning
to the mundane, however, we must perceive that we are, as national and international
societies, going through a series of "Industrial Revolutions"
that follow closely the sequence of biological progressions.Several
millennia ago we learned to communicate verbally and in writing and to use
tools. These were remarkable revolutionary advances.In
the middle of the last century we learned the biomimetic trick of harnessing
external power sources in the way we had learned to use animals to provide
extension, speed and intensification of our own muscular power generations.
We
also learned rudimentary telecommunication, that is an extension of our
voice, our writing, our lever pushing to a remote point. This is the generally
accepted "first" industrial revolution.These
biomimetic societal discoveries do not shut off when a new one starts any
more than infant learning abilities stop abruptly when juvenile windows
open. It is hardly realized by most engineers that while James Watt and
Joseph Black were setting up their half century partnership to build better
steam engines while developing the theory of specific heat and latent heat,
chemists like Kehule and Liebig were demonstrating the feasibility of artificially
growing desired kinds of molecules. This molecular agriculture slowly built
until about World War I, when synthetic organic and inorganic chemistry
were big and growing industries. [unreadable.] Only within the last few
years have we addressed this biomimetic process more directly as we tackle
the technical, legal and ethical problems of more blatantly bioengineered
molecular synthesis under the buzzword "genetic engineering".
Starting about
1926 we entered the higher-level control and communication revolution which
did not really flourish until the end of World War II. It is just hitting
its stride now and can be expected to advance for several more decades.
But like a baby
who has learned how to flex its muscles and has acquired the skill needed
to look at, listen to and feel its surroundings, but has only primative
ideas of what to do with these skills, we are just preparing to engineer
intelligent and intimately cooperative [unreadable] systems that will work
at the management, the supervisory, and we expect at the intellectual levels.
Those of us under twenty will stand a good chance of participating in this
new [unreadable] type engineering.But
now let us look at some name-calling. Without a name around which to rally,
a science will languish. Biomimetic science is a hard one to name because
there is an ego-loss involved so soon as a once in a lifetime inventor or
innovator admits that he is simply freeloading on biological nature.As
a brash kid, I tried tackling this field for a Ph.D. thesis, combining majors
in Physics, Biology and Mathematics to do a biomimetic simulation of the
nerve axon and capitalizing on the engineering insights of building a thing
called "an iterative analog computer" at the time when computers
were not in the common jargon. I was handed a Ph.D. examining committee
of eighteen members including practically the senior statesman of physics
[unreadable] and the department of engineering and the [unreadable] and
"almost failed to make it because of battle fatigue before each of
these was done with me. Some
years later, in 1954, I tried to sell the name "Biological Engineering"
for this biomimetic process. The circumstances of this effort are most amusing.
My brother had just accepted at that time the position as head of the department
of Biology and Biological Engineering at MIT with a strict stipulation that
the Biological Engineering part be firmly deleted from the name.
He didn't want messy practical-minded engineers cluttering up his idealistically
conceived department delving into macromolecular biology. He developed a
fine, productive department that could never be accused of pursuing practical
engineering goals, but to tease him I adopted the name Biological Engineering
for Biomimetic engineering and published same articles under this title
like that in Honeywell Flight Lines [magazine]. The name never sold.Norbert
Wiener, with much the same idea in mind, had published his book Cybernetics
in 1946. Grey Walter claims that Ampere had used the term cybernetique a
century before to describe a technologically organized form of government.
From conversations with Wiener and with Arturo Rosenblueth and Warren McCullock,
who supplied him with much of the biological, physiological and psychological
information on which he based his book, I am sure that he wanted Cybernetics
to mean a much broader biomimetic concept than simply that of ordinary information
and control theory, yet Cybernetics lives and occupies that niche.Major
Steele, with the U.S. Air Force at Wright Patterson Air Force Base, tried
to introduce Bionics for this science of design based on life systems. His
term lives, but mainly in the comic sections and science fiction TV programs
as assorted Bionic women, men, and dogs.Robotics
is a possible contender for the Biomimetics contest. It's a good buzzword
now, and although usually confined to anthropomorphic skillful but rather
stupid manipulations, could prevail as CAD-CAM moves on to CIAM or ICAM
and allows robots to think. [Unreadable]Biomimetics
is easy to say, is etymologically correct, and has few bad connotations,
but does not have market appeal - at least not yet.Perhaps
this problem of popularizing biomimetic engineering could be taken to the
uninhibited youthful engineering group in much the same way that invention
of computer implemented aids for the handicapped was done last year in the
Johns Hopkins Applied Physics Laboratory effort in which I participated.
That has already yielded a fat special publication in the IEEE Monograph
seriesBefore we
get to the fun thing of examining cases in point in each of the three Biomimetic
categories, let me come to the final point.Can
we and should we rearrange our educational and our technical societies and
academies so that we can produce individually packaged and custom educated
multidisciplinary engineers and scientists without requiring that each survive
the entire traditional program for a standard prototype mechanical, electrical,
computer, chemical, civil, medical or what not engineer? Even with our much
complained-about auto production industry, we can still get a Custom made
car with desired finish, fittings, power controls, engine, transmission
and Interior decor. The delay is only a couple of months.Why
can't we offer a bright student a career training combining mechanical engineering,
electrical engineering, control science and enough business and marketing
experience to allow him to design clever new robots or alternatively to
develop surgically implantable computers and controls to aid us as impaired
or even as nominally impaired humans?There
are a gleams of hope in several directions. I
have already mentioned the Minnesota Innovation Center as a route toward
tripartite cooperation in developing problem-oriented designs for education,
regulation, marketing, etc. There is a Minnesota technology transfer committee
that has been working toward these same ends and is now being reorganized
for more aggressive action.Another
unusual opportunity presents itself for renewing the spirit that was in
our organizational design for the Biophysical Society, just a quarter century
ago, as an umbrella organization to bridge and unify the Biophysical Sciences
and Technologies, basic and applied, and in the Alliance for Biomedical
Engineering.For
the first time in its history IEEE, the largest engineering organization
in the world, is actually holding a national election with [unreadable]
options for each office, including the presidency. One of my old friends
in Bioengineering, Dick Cowan, is running for that office and has asked
me to contribute ideas for viable, meritorious campaign issues on which
to run and projects to be initiated. Just possibly Professor Cowan, now
Dean of Engineering and vice-president of South Dakota U. could sell the
idea of a biomimetically based umbrella cooperation between IEEE, with its
several dozen constituent societies and its sister societies in the engineering
fields, and in basic sciences such as ASME, ASCEE, AIP, AIBS, ACM etc.To
show you that Major Cowan cannot be all bad, let me show you the collection
of "New Math" that he gathered from his bioengineering student
examination while a professor at the Air Force Academy and sent out as a
Christmas card. Now,
as a final entertainment, I thought it would be interesting to list, simply
as they come to mind, a hundred or so personal inventions, good, bad, and
indifferent, distributed among the three categories. Most of these have
biomimetic roots and it will be productive now or later to see how these
basic principles emerged and where they have gone.It
may interest you to know that among other rather sweeping changes that have
happened or are being proposed for the U.S. Patent Office and its auxiliaries,
there is a movement toward adding an algorithmic area of patentability.
This seeks a compromise between the traditional rule against patenting a
"law of nature" and the very legitimate patentability of a new
and useful restatement of such a law in a conveniently understandable and
maniputable form.[Unreadable]As
a first case, consider the Heat Pipe now in considerable use, which was
published under the heading "Vapor Cooled Electrodes". It follows
the old panting dog strategy of drinking water and then blowing it off as
vapor to use the heat of vaporization as a refrigeration medium. The same
or other water is then drunk in liquid form to make up the deficit and recycle
the working fluid.Another
widely used technique, published under the title "Electrical Control
of Galvanometer Characteristics", recognized that by differentiation
of a position variable of a movable machine component one can get measures
of x, x', x'' or 0, 0', 0'' depending upon whether linear or angular motions
are involved. Each of these, suitably multiplied by a constant and applied
through a force-torque transducer, can effectively imitate negative or positive
mass, Moment of Inertia, compliance - restoring spring constant or damping.
So that the device as seen from external tests would have whatever characteristics
of damping, frequency response, sensitivity that are desired. This is in
direct imitation of some biological active feedback controls. The technique
has been widely used everywhere from ship steering to loudspeaker frequency
supposition to machine tool controls.The
Schmitt Trigger circuit developed specifically to simulate the nerve axon
behavior, has become universally accepted as the route to bistable hysteretic
control of systems that want to dither or chatter about a servo set point.A
relative of these last two cases is one that is deserving of greater attention
than it has received. This is the three quadrature compaction control algorithm. |
