What follows is an intro paperwriten by Mark Tilden. Although it is informative and describes the sciencebehind Nervous nets It is writen in Academic Speak. So for those who ratherget the layman's version tryhere .

Biomorphic Robotics and NervousNet Research:
A New Machine ControlParadigm

Mark W. Tilden,

Biophysics Division,
Los Alalamos National Labs

Submitted for publication to the EANN '95 Conference Proceedings"Special Track on Robotics."

Nervous Net (Nv) technology is a non-linearanalog control system that solves real time control problems normally quitedifficult to handle with digital methods. Nervous nets are to Neural netsthe same way peripheral spinal systems are to the brain. This work hasconcentrated on the development of Nv based robot mechanisms with electronicapproximations of biologic autonomic and somatic systems. It has been demonstratedthat these systems, when fed back onto themselves rather than through acomputer-based pattern generator, can successfully mimic many of the attributesnormally attributed to lower biological organisms. Using Nv nets, highlysuccessful legged robot mechanisms have been demonstrated which can negotiateterrains of inordinate difficulty for wheeled or tracked machines. Thatnon-linear systems can provide this degree of control is not so surprisingas the part counts for successful Nv designs. Afully adept insect-walker, for example, can be fully controlled and operatedwith as little as 12 standard transistor elements.

Since the start of research in the winter of1994, development of this technology has advanced to solving currentlydifficult sensory and cognitive problems. It is hoped in the coming yearsNv systems may do for robot vision (amongst other disciplines) what hasbeen done for autonomous robot vehicles, namely the reduction of currentlycomplex systems down to an inexpensive but robust minimum. Further effortsare also being made to apply this control strategy to the expanding nanotechnologyfield. At the nanometer scale Nv's may prove more feasible than nano-computersfor control of self-assembling micro structures. For now, however, Nv researchconcentrates on problems of scale invariance, proving by example (or exhaustion)this control system can work at all scales, types, and styles of roboticapplication.

The Nv control method could be adapted to mosttypes of machine control, but it has been applied to autonomous robotsbecause of the difficulty conventional control systems have solving theseemingly simple task of negotiating undefined complex environments. The80 or so "biomorphic" robots (from the terms BIOlogy and MORPHology,and the Latin for "living" and "form") built so farare not "workers" in the traditional sense, but "survivors",in that they fight to solve the immediate problems of existence ratherthan procedural condition (i.e.: they do not follow the rules of an internalprogram that mimics the external world, but the world itself). Nv controlarchitectures focus on adaptive survival rather than the performance ofspecific tasks. Once survivability is under control, goals can be superimposedand the machine used as a platform to carry sensors and conventional electronicintelligence. It is believed that these machines, although now in an earlystage of development, can within a few years be brought to the point thatthey can serve as inexpensive, robust, and versatile carriers for a varietyof instruments. A vast number of applicationswould then be possible, including the location and possible clearing ofland mines from civilian areas, security,maintenance, medical and prosthetic applications (a cost-effective "walkingwheelchair" for example), and even cars with on board "survival"instincts to save themselves, and their passengers, from damaging accidents.Though the Nv based legged devices built so far cannot go everywhere, theycan certainly go places not currently accessible to wheeled or trackedvehicles of similar scale. It seems that for handling undefined environments,biomorphic designs are a very efficient and cost-effective approach.

Initially it was thought these devices avoidedthe problems of an internal world representation by using a reactive orbehavior-based technique. Recent work hasshown however that Nv biomorphs instead take a chaotic map of their surroundingsonto their process control hierarchy (that is, they dynamically and efficientlyadapt to the fractal complexity of their surroundings).This is due to the analog-electronic nature of the devices, the adaptivehardware of their structure, and the topological orientation of their interconnections.The defining characteristic of this adaptation is continuously updatedby the immediate fractal complexity of the environment. These devices are"soft" designs, in that the environmental dimension must be absorbed,modified, and acted on for the devices to make successful headway througha complex world. These devices do not use"feedback" in the standard sense, but rather "implex",as the driving forces are augmented by perceived load rather than by aseparate regulating path. The result is highly compliant, animal-like machinemotions that "negotiate" rather than "bully" theirway through environments, resulting in minimal damage to both world androbot.

We talk about these devices in the generalsense because the precepts of their existence and subsequent design arebased upon environmental macros, such as fluidity, turgidity, gravity,scale, materials strength, and many other factors. The power of biomorphicdesigns is that this information is used as the defining principles toshape appropriate survivor(s) for a particular environment. The machinesthat emerge are vastly different from any conventional robotic forms. Wesuspect, at least from the experimental evidence, that this technique embodiesa new type of non-linear control paradigm, and at least an entirely newengineering discipline for the matching of competent machines to complexenvironments. Here, once the problems of existenceare ratified, the devices can do unsupervised, long term work without humanintervention (some devices have been in continuous operation for over 5years).

The potential for this control paradigm isvast, but it is far from linear, and requires integrated design attemptsto pull a competent ability from the Nv nets.To this end, the use of this technology to "evolve" machinesfrom a lesser to a higher operational state hasresulted in not only a wide spectrum of devices, but even completely different"species" of creatures, all evolved from a primal "genotype";the single "cell" creature known as Turbot 1.0.

A further advantage is the speed at which thisevolution has occurred, indicating that real-world Lamarckian evolutionmay match the success of many computer models yet seen. The diversity ofthis technique offers potential solutions to two main research fields,macro and micro robotics, and experimental work has been done to produceadept prototypes for both. The conclusions are that there may be some universalchaos-bounded concepts that bind survivor oriented designs, allowing forthe creation and optimization of devices that can do work in any environment,under many situations, using chemically inert, and thereby relatively safe,control techniques (the idea of seeding a wheat field with pest-fightingsilicon biomorphs to produce high yield, insecticide-free foodstuffs isan attractive example).

Considering that biomorphs may last long enoughto replace most forms of long-term damaging chemicals (i.e.: pesticides,bleaches, medicines) the potential for the field really opens up. Deployedartificial chemicals perform a task in their immediate area of concentration,and then disperse into the environment where, after a time, they cannotbe absorbed adequately. Biomorph machines, made from biodegradable siliconand trace elements, can be made gregarious so they do not of their on volitiondisperse, and can be absolutely controlled by conventional methods. Whetherat micro or macro scales, these designs are not just capable, but competent.Furthermore, as they are self "programming" and non-reproductive,their behavior is both contained and predictable.

Nv biomorphs are something new with a demonstrablepotential. Future work will concentrate on how this technology could fillin the cracks between science fiction and reality by finding out what isfeasible now, and how to logically proceed to marketable, capable machines.In the coming years, it is hoped to be possible to demonstrate real machinesto assess feasibility for macro and nano robotic applications. Expansionsof the fields of robobiology, robomorphology, and artificial ecologys willbe studied and published, along with extensions of this field from self-repairingprocessors, new computational paradigms, and even nanorobotic surgeons-in-a-capsule.Biomorphics is new, but it is slowly gaining the maturity and acceptancenecessary to become a valid work tool.