Replicating Automata

Ideas of automatic machines and robots have existed for centuries, but not till the work of John von Neumann in the mid twentieth century was there mathematical proof of the possible existence of self-replicating automata in a real physical sense. Assuming the proper material could be found, and using relatively simple neighbor logic, subunits could spontaneously assemble into complex structures, dynamically disassemble, multiply, or rearrange into successive configurations. Freeman Dyson (1979), Schneiker (1986) and others have considered the profound usefulness of such hypothetical robotic replicators in uses and sizes ranging from outer space exploration to household furniture to nanoscale "Hibb's machines," or nanodoctors. Dyson (1979) foresaw self-replicators which, after being launched from earth to an asteroid or planet, could mine, their own raw materials and form symbiotic relationships with other "life" forms. Shoulders (1961) extended the concept of replicators to the nanoscale where their existence could have wide ranges of scientific and medical applications.

Implementation of real-world, useful replicators faces many obstacles. A NASA study group (Bekey and Naugle, 1980) found that von Neumann's famous proof made major assumptions and avoided significant problems. Schneiker (1987) has summarized an approach to simplifying the development of replicative automata by minimizing the different types of materials/parts needed, the number of scale-dependent factors, part tolerance and wear, and assembly complexity. One other requirement is to increase reliability (i.e. by intrinsic error detection, repair, regeneration). Presently, these traits are fulfilled only by computer simulations or purely biological structures such as cytoskeletal proteins and viruses.

Schneiker (1986) notes that simple microreplicators, augmented with STM/FMs could mass produce nanotechnology products in virtually unlimited quantities. Nanotechnology applied to new superconductive materials (and vice versa) may help to implement useful replicative micro-automata which in turn could turn out nanodevices in vast quantities. Until recently superconductivity has been thought limited to near absolute zero temperatures, however some materials have been shown to have superconductive transitions at much warmer, easily attainable temperatures (Robinson, 1987).

The dramatic increase in superconducting transition temperatures in certain materials, and the availability of lossless current, magnets, and motors may herald a wave of scientific, technological, and economic advances. Schneiker contends that among these will be the facilitation of STM technology (superconductive tunneling), Josephson junctions, magnetic field sensors, as well as superconducting replicators. Superconducting materials offer scaling benefits, design simplicity, low power needs, and high component density. Superconducting magnets and coils could form switchable "glue" or "clamps" to hold subassemblies, generate switchable magnetic field patterns, direct assembly, positioning, and dynamic functions, drive rotary/linear motors or form superconducting channels or guideways for material transport (Schneiker, 1987).

Collective intelligence occurring in parallel replicating automata has been studied by Christopher Langton (1986) of Los Alamos National Laboratories. Langton is the organizer of a Los Alamos conference on "Artificial Life," the study of computer systems that exhibit behavior characteristic of natural life. By exploring models of cellular automata and self-replicators, Langton (1987) hopes to "extract the molecular logic of living systems."

Langton (1987) states.

Microelectronic technology and genetic engineering will soon give us the capability to create new life forms in silico as well as in vitro. This capacity will present humanity with the most far-reaching technical, theoretical, and ethical challenges it has ever confronted.

0 0

Post a comment