LESSON_11: ARTIFICIAL BEHAVIOUR
<http://www.unm.edu/~amri>
That link gives you a home page with one of the most dramatic visual displays you will find on the web. A human heart is beating against the backdrop of a starry sky to announce the "Historic FIRST WORLD CONGRESS ON BIOMIMETICS AND ARTIFICIAL MUSCLE", held in Albuquerque, Dec. 9-11, 2002.

A Google search using the key words, "artificial muscle" yielded 269,000 references.  This field may dramatically change the society in which we live overnight if a breakthrough is made, launching us into what will unmistakeably be a science fiction-like era for the entire human race.

Succintly what it amounts to is this. An artificial skeleton, much stronger than the human skeleton can obviously be made now. What if "electroactive elastomers" (which yielded 320 Google references) could be placed over that skeleton and artificially controlled to evoke the range of overt behaviours of a human ... and then some?

Such a creation would have an immediate market even if it could do only the simplest tasks of a normal adult human. The television series "Total Recall: 2070" has a humanoid robot like this, set in the year 2070. Given the analysis so far concerning SHAIs, the prospect of a sudden breakthrough which give us a "Farv" from the Total Recall series could happen before this decade has ended.
HISTORY AND CURRENT SURVEY

A University of Dortmund site gives us some history under the title, "Walking Machines: 2-legged-robots". Waseda University is credited with the first two legged robot to walk like a human, ie with "dynamic" walking, in 1980.
<http://lls11-www.informatik.uni-dortmund.de/people/ziegler/RoboterLaufen2Legs.html>

For legged robot sites around the world, see:
<http://www.kimura.is.uec.ac.jp/faculties/legged-robots.html>
AN EXCITING CHALLENGE

Tom Edwards has an honors degree in mechanical engineering we are told on his web site for AECOMR, the All Environmental Center of Mass Robot. This is an exciting and innovative approach to the problem of legged locomotion and should be taken seriously (though as always, critically) because of the claims made and the credentials of the person making the claims. He observes that "...very few robot designers ever moved the center of mass of their robots ... almost all animals even insects moved their center of mass". AECOMR has two wheeled legs and a body or center of mass which can move along a rail connecting them. By shifting the center of mass (Center of Gravity or Zero Moment Point) Tom claims his robot "...can go anywhere a human can bar climbing a tree".

That also sums of the challenge of legged locomotion. Can a robot be made which will go anywhere a human can goes by using legs ... and further, does it have "human equivalency"? Again we can draw a profile for human equivalency as there are different manifestations of legged locomotion. Can it kick a football as well as a human? Climb stairs as well as a human? Can it walk and run as well as a human on level ground? Some day we will see contests which incorporate these aspects of legged locomotion. The first robot sprinter to beat the world's fastest human in the 100 yard dash will be quite an event for the history books.

Click on the kimura web site above, go to the Leg-Wheeled section and click on "Center of Mass Robot"
While mechanical engineers presently dominate advances in robot locomotion, giving us such exciting prototpes as Honda's Asimo (Japanese for "legs"), we return to the elastomer-over-skeleton robot for the final word in this lesson. We read in the article "Biological Inspiration For Muscle Like Actuators" by Kenneth Meijer that "WE ARE AT THE BRINK OF A BREAKTHROUGH IN THE DEVELOPMENT OF LEGGED MOBILITY PLATFORMS WITH BIOLOGICAL PERFORMANCE".

What might justify such a statement? The justification is found when we read that these electro-active elastomers have  "...capacities within the ranges reported for natural muscle. They can generate similar forces and strains ... their capacity for doing work and generating power
MATCHES THAT OF NATURAL MUSCLE". (Blocks are mine).

If that is correct, the breakthrough will then come from precise electro-active control of these artificial muscles. For a more general treatment of this field at JPL from which the Meijer paper originates, google on JPL artificial muscle. If JPL or another center should make the breakthrough in elastomer control. not only will the $100 m. Asimo be obsolete before it reaches the mass market, but
OUR ENTIRE CIVILIZATION WILL BE TRANSFORMED. Even with the AI we can muster now and such humanoids doing only the most menial of tasks, the world will be a very different place if ...when ...they arrive.



Hosted by www.Geocities.ws

1