Advanced Automation for Space Missions/Chapter 4.5

4.5 Automation and Manufacturing Technology Requirements

To realize the full potential of space manufacturing, a variety of technological development programs should be initiated in the near future. It is strongly recommended that NASA focus research attention on improvements in teleoperation and robotics, automated manufacturing techniques, and advanced materials processing.

Space manufacturing efforts will draw heavily on teleoperation at first, gradually evolving over many decades towards the extensive use of autonomous robots. Additional research in teleoperation is needed immediately on sensors - tactile, force, and visual, and on sensor and master-slave range scaling. Robotics requirements include improvements in decisionmaking and modeling capabilities, sensors and sensor scaling, mobility, adaptability to hazardous conditions and teleoperator safety (Schraft et at, 1980), natural language comprehension, and pattern recognition. Many of these needs are presently under review by the Engineering Services Division of Goddard Space Flight Center as part of their ongoing CAD/CAM program.

Better automated control systems for space-manufacturing processes are imperative. Machine intelligence controlled laser-, electron-, and ion-beam technologies will make possible the highly sophisticated cutting and trimming operations, integrated circuit fabrication, and other related functions necessary for an efficient SMF operation. Further work should be aimed at devising new fabrication techniques specifically designed for space, such as automated beam builders.

In the materials processing area, effective use of undifferentiated materials such as cast basalt should be stressed. Beneficiation systems better suited to nonterrestrial conditions must be developed to achieve production of differentiated materials with maximum process closure.

4.5.1 Teleoperation and Robotics

Teleoperator development is especially important in the early stages of the space manufacturing effort because the sophistication of current robots in sensory scaling, adaptive control, learning, and pattern recognition is inadequate to establish an autonomous space manufacturing capability. These skills are embodied as subconscious processes in the human nervous system. The development of teleoperators with sufficient interface dynamics would provide "telepresence" (Minsky, 1979, 1980) in the early stages of SMF development while significant new robotics research is undertaken.

The team surmises that within the next 50 years robot systems will be capable of handling a large fraction of the needs of a general-purpose SMF. The feasibility of robot systems making sophisticated judgments is less certain. Controls likely will evolve from teleoperated to semiautomated, then to fully automated (Bejczy, 1980). Cost requirements in orbit or on the Moon or asteroids may encourage development of adaptive robots with flexible control systems (Asada and Hanafusa, 1980). According to research currently underway at the School of Electrical Engineering at Purdue University, a limiting requirement may be manipulator motion (Paul et al., 1980). Manipulators in an SMF must be capable of working on a moving assembly line the maximum "reach" of current Cyro robots is 3 m - and or accepting visual position information. It is also important to determine the degree to which real time computational constraints can be relaxed in controlling robot motions in Cartesian coordinates. In extraterrestrial environments, the dynamic behavior of each link in a manipulator arm must be considered. Centrifugal and coriolis accelerations (in spinning systems) and gravity loading are significant factors governing the relationship between forces and moments of successive links.

Limits on control requirements also have been considered by Yushchenko (1980), who has written algorithms for semiautomatic robot operations. Since semiautomatic robots undoubtedly v/ill precede fully automatic robots into space, the three major techniques of direct human master control - velocity, force, or position - must be considered. Velocity methods are rapid but manipulator motions are imprecise. Force methods control manipulators through human feedback in Yushchenko's study, but these techniques provide little regulation of acceleration during object motion. Limitations in force-sensing controls for mating of parts have been reviewed by Korolev et al. (1980) and by the Draper Laboratories, the latter quantifying clearance and friction factors. The positional method ensures proportionality of linear and angular displacements of manipulator grip through the handle of a master control device.

Manipulators need to be greatly improved. Current master-slave devices require 2-3 times longer to accomplish a given task than do human hands (Bradley, personal communication, 1980). The mass of teleoperator appendages is high compared to the weight they can lift. With better visual and tactile feedback, the heavy, rigid manipulator arms could be replaced by lightweight, compliant, yet strong arms. To accomplish this, the low-resolution, low-stability, low-dynamic-range force reflection tactile systems must be replaced with servofeedback systems including suitable touch display modules. Viewing systems will require additional research and development - the most advanced system currently available is a monocular head-aimed television. This system should be redesigned as a binocular system with auto-focus, variable resolution, and color. Sensory scaling to compensate for differences in size between slave and master manipulators is necessary for fault-tolerant teleoperation. This may be accomplished by adjusting the scale of the master visual image or by incorporating error signals into the visual display.

Limitations also arise by virtue of the space environment itself, whether in LEO, on the lunar surface, or on asteroids. Hard vacuum demands redesign of robot joints and manipulator end-effectors to minimize undesired cold welding if de-poisoning of metal surfaces occurs. Radiation bursts during solar flares could possibly induce embrittlement of metal components of automata. Likewise, electronic components could be degraded or altered by temperature extremes.

4.5.2 Functional Requirements for Automation

The functional requirements for an automated SMF, taken in part from Freitas (1980d), are listed below roughly in order of increasingly sophisticated capability: robot language systems, product assembly, product inspection and quality control, product modification, product repair, product adjustment, product improvement; remedial action by reason of emergency or subtle hazard, robot self-replication. It is assumed in each case that the impediments to meeting these requirements (e.g., control techniques, "packaging" to withstand hostile ambient environments, etc.) will somehow be overcome. The first three functional requirements are described briefly below, followed by a general discussion of the more advanced requirements.

Robot control languages. Numerous machine languages exist for the control of semiautomated machine tools (Lindberg, 1977). These include APT (automatic programming tool) and ICAM (integrated computer aided manufacturing). McDonnell Douglas Aircraft Company has recently extended APT to MCL (manufacturing control language) in order to program a Cincinnati Milacron T3 robot to rivet sheet metal. Higher-level robot control languages, obvious requirements for advanced automated space systems, include VAL (versatile assembly language) for the Puma robot and "HELP" for the Pragmac robot (Donata and Camera, 1980). The problem of extending high-level languages from comparatively simple machine tools to more sophisticated multiaxis integrated robot systems which may be found in future automated space factories must be viewed as a top priority research item.

Product assembly. At SRI International, requirements for the five basic operations in factory assembly have been evaluated by Rosen et al. (1976). These include (1) bin picking, (2) servoing with visual feedback, (3) sensor-controlled manipulation, (4) training aids, and (5) manipulator path control.

The team has recognized the need for improved performance in bin picking of, say,assorted cast basalt and metal objects. Multiple electromagnetic end-effectors certainly could pick out just the metal casings. Variably energized end-effectors might be used to separate and select metal parts of varying magnetic susceptibility randomly arranged in a bin (i.e., aluminum vs iron vs titanium parts). But general bin picking from random parts assortments is not yet possible, though it might be essential in a fully automated SMF operation.

SRI has applied visual servoing by combining a General Electric television (100 X 100 element solid-state) camera with an air-powered bolt driver incorporated into an end effector. Three-dimensional cameras may be required for highly contoured objects fabricated in space (Agin, 1980; Yachida and Tsuji, 1980). Such cameras have already been applied to automated bin selection tasks by the Solid Photography Company in Melville, New York.

Computer-vision technology needs to be merged with discoveries from biological studies. Automatic gain control, gray-scale imaging, and feature detection must be included in computer-vision technology if robot autonomy is the goal. Parallel computer-control systems will ensure the speed of reaction and self-preservation "instincts" required for truly autonomous robots, but will require a decrease in existing computer memories both in size and access time by several orders of magnitude. Consideration should be given to associate and parallel memories to couple perceptions to the knowledge base in real time.

To achieve sensor-controlled manipulation, somewhat greater precision Is required of robot arms than can be obtained now. Present-day Unimates (control and precision of 2.5 mm) have been used in a one-sided riveting operation using strain-gauge sensing of the rivet gun mandrel, but there is still a need for more rapid finding, insertion, and fastening by passive accommodation, servo adjustment, and search algorithms. A novel "eye-in-the-hand" adaptation for rapid assembly in space may utilize acoustic sensors. The Polaroid Corporation in 1980 applied its camera ranger to end-effectors for tool proximity sensing. The unit emits a millisecond pulse consisting of four ultrasonic frequencies (50, 53, 57, and 60 kHz). Ultrasonic techniques are potentially quite useful in air or other fluid-filled bays in nonterrestrial manufacturing facilities, especially in view of the acoustic positioning systems developed by the Jet Propulsion Laboratory for containerless melt manipulation. Under vacuum conditions when precise positioning is necessary, laser interferometry may provide the answer (Barlunann, 1980).

Regarding training aids, more sophisticated coordinate transformation programs are required to operate manipulators for diverse tasks. A possibility for the future is "show and tell," a new technique for robot training (see chapter 6). Ultimately, a robot itself could train future-generation machines through some means of "training-by-doing." A related issue - the problem of robot obsolescence - will not be trivial.

Finally, manipulator path control should be fully automated in SMF where, for example, rock melts must be transported along smoothly controlled paths (see the discussion of basalt fiber spinning in section 4.2.2). In the manufacture of bearings or fibers where high-speed trajectories are involved, manipulator halts at corners must be avoided by developing better path control strategies. In the near-term, it may be possible to extend the capabilities of the Unimate:PDP-11/40 couple. For every machine proposed for the SMF, including the starting kit extruder, it is simplest to use a coordinate system based on that machine to interact with robot manipulators continuously to redefine forbidden regions and motions. Thus, a major requirement in robot factory assembly is to specify the coordinate systems of the component machines.

Product inspection and quality control. The need for visual methods of inspection and quality control by automata must be defined for each class of SMF product envisioned. For instance, the application of electroforming on the Moon to produce thin-walled fragile shapes, aluminum ribbon extrusion, or internal milling of Shuttle tanks, definitely demands inspection and quality control. Terrestrial automated inspection systems currently are in use at General Motors, Western Electric, General Electric, Lockheed Recognition Systems, Hitachi Corporation, SRI International, and Auto-Place Corporation. A detailed synthesis of the vision requirements for each is given by Van der Brug and Naget (1979). Off-the-shelf television systems with potential for robotics applications already provide measurements to 1 part in 1000 of the height of the TV image, e.g., the EyeCom Automated Parts Measurement System manufactured by Special Data Systems, Inc. in Goleta, California. Finally, the use of fiber optics in quality control, as demonstrated by Systems now in use by Galileo Electronics, Inc., warrants further development.

Advanced functions and recommendations The needs of space manufacturing for automated product modification repair, adjustment and improvement, as well as robot adaptation to emergencies and self-replication, depend in large part on the capabilities of future automata control system and the environment in which they are applied. The hazards of space to human beings are well known, whereas the impact on robot systems is less well understood. Potential dangers include rapid pressure changes, spillage of corrosive fluids or hot melts due to vessel rupture, radiation effects from solar flares (e.g., embrittlement), anomalous orbital accelerative perturbations producing force-sensor errors, and illumination-intensity variations caused by space platform tumbling or nutation (producing visual observation problems such as shadow effects in fiber optics sensors).

Robotic intelligence must be vastly increased if these devices are largely to supplant human workers in space. This may be accomplished by deploying a versatile intelligent multipurpose robot or by developing a number of specialized, fixed-action-pattern machines. Multipurpose intelligent robots lie well beyond state-of-the-art robotics technology, yet they still are an important ultimate goal. In the interim, sophisticated fixed-action-pattern robots suitable for restricted task scenarios should be developed. The behavior of such robots would be not entirely different from that of many plants and animals endowed with very sophisticated fixed action patterns or instincts.

Before true machine intelligence can be applied to factories in space, the requirements for automated nonterrestrial manufacturing systems must be determined by an evaluation of the state-of-the-art in this field. A complete and updated computerized library containing abstracts of all available robotics research and applications publications, accessible through ARPANET, should be implemented to enhance automation technology transfer. Among the subject categories which should be emphasized are controls, arm/work envelopes, robot adaptability, applications, and costs. Knowledgeability in the field requires contact with finns listed below to better understand how solutions of the practical problems of today can be extrapolated to help solve those of tomorrow: Unimation, Inc.; Cincinnati Milacron; ASEA, Inc.; Prab Conveyors, Inc.; Planet Corporation; Devilbiss/Trallfa; Nordson Corporation; Binks, Inc.; Thermwood Machinery Corporation; Production Automation Corporation; AutoPlace Company; Modular Machine Company; Seiko Instruments, Inc.;Jones Oglaend Corporation; Fujitsu Fanuc Corporation; Okuma Machinery Corporation; Advanced Robotics Corporation; Hitachi Corporation;and Benson-Varian Corporation.

4.5.3 Space Manufacturing Technology Drivers

The successful deployment of a large, growing, independent SMF requires technologies not presently available. Three technical areas in particular will require major developmental efforts: manufacturing technologies, materials processing, and space deployment. Many of the technology drivers and required advancements discussed previously are currently the subject of some R&D activity at various industrial and government research facilities. The first and perhaps most crucial step in any technology drive to make the SMF a reality is a thorough synthesis and coordination of current and previous research. A determined effort must then be made to augment technical competence as required to sustain a successful space manufacturing venture.

Manufacturing technologies. The control system for an automated manufacturing facility must be sophisticated, fault tolerant, and adaptive. Technological advances required for a factory control system are primarily software developments. A "world model" for the facility must comprehend variable throughput rates, breakdowns, and unexpected commands from Earth-based supervisors. The control system also must be able to formulate and execute repair plans, retooling exercises, and scheduling options. Such a system needs flexible hypothesis formation and testing capabilities, which in turn demands heuristic programming employing some measure of abductive reasoning without requiring unreasonably large memory capacities (see sec. 3.3).

Advances in ion-, electron-, and laser-beam technologies are necessary for welding, cutting, sintering, and the fabrication of electronic components. The efficiency and power of weapons-grade tunable lasers now under development by Department of Defense contractors (Robinson and Klass, 1980) already are high enough to fulfill most cutting and sintering needs of the SMF. Heat dissipation is a substantial problem inherent in laser utilization for space manufacturing. Space-qualified heat exchangers must be developed for laser-beam machining to achieve its full potential as a viable macromachining space technology. In addition, industrial lasers must be designed to re-use the working gases.

In the manufacture of electronics components, ion-beam devices are required for implantation and etching in space. Lasers are helpful in facilitating annealing and oxidation processes and in trimming fine-tolerance capacitors and resistors. Electron beams have applications in silicon crystal purification and deposition of metals, though lasers also may be employed. Other uses for each beam type are readily imaginable. High-resolution automated control technologies must be developed for implantation, annealing, etching, and trimming processes in particular.

Contact welding is a highly useful feature of the vacuum space environment. Of course, in some instances cold welding must be avoided so surface poisoning methods must be developed. Terrestrial poisoning agents such as hydrogen. hydroxyl, and various surfactants are not readily produced from nonterrestrial materials. Highly adsorptive oxygen-based surface active agents appear to be the most feasible solution to the cold welding problem.

Materials processing. Extensive research is needed in the field of processing of raw materials if a self-sufficient manufacturing presence is to be established. Several possible avenues include fractionation, zone refining, and oxygen-based chemical processing. Fractionation of a wide variety of elements including fluorine, hydrogen, silicon, boron, phosphorus, and many others is a prerequisite to independent manufacturing in space. Raw material separation prior to processing (primary beneficiation) is a logical step in the total beneficiation process. The preliminary isolation of particular compounds or mineral species could significantly reduce the problems inherent in developing suitable chemical-processing options.

Space deployment. There are a number of mission tasks associated with space manufacturing for which technological developments must be made. Sophisticated rendezvous techniques are needed for SMF resupply, in-orbit assembly, and satellite tending. Deployment of repair rovers is required for satellite maintenance and troubleshooting. Long-term satellite autonomy is not possible without repair and refueling capabilities which are not currently available. Large-mass deployment and retrieval procedures must likewise be developed if feedstock, raw materials, and products are to be delivered to or from the SMF. Multimission compatibility must be designed into satellites, shuttles, and transport vehicles if self-sufficiency is to be achieved within a reasonable time.

4.5.4 Generalized Space Processing and Manufacturing

A generalized paradigm for space industrialization is presented in figure 4.20. Solar energy powers the systems which gather nonterrestrial materials for conversion into refined materials products. These "products" can be additional power systems, materials gathering/processing/ manufacturing systems, or simply support for other human and machine systems in space. Earlier chapters examined observational satellites for Earth and exploration systems for Titan having many necessary features of a generalized autonomous robotic system designed to explore the solid and fluid resources of the Solar System (item (1) in fig. 4.20) using machine intelligence. However, in the materials and manufacturing sectors a qualitatively new interface must be recognized because "observations" explicitly are intended to precede a change of objects of inquiry into new forms or arrangements. These machine intelligence systems continuously embody new variety into matter in such a way that preconceived human and machine needs are satisfied. This "intelligently dynamic interface" may be explored as two separate notions: (1) a generalized scheme for materials extraction, and (2) the (fundamentally different) generalized process of manufacturing (see also chap. 5).



Generalized materials processing system. Figures 4.21 and 4.22, developed by R. D. Waldron (Criswell, 1979), offer a very generalized overview of the options and logic involved in the selection of a processing system for an arbitrary raw material input. By way of illustration, note that the extraction (in either reduced or oxide form) of the seven most common elements found in lunar soils requires at least six separation steps, with yet additional steps for reagent recycling. Even if a single separation technique from each of the 22 categories shown in figure 4.21 Is considered for each of the six lunar elements, more than 113,000,000 combinations (22$6$) of separation would be possible. The 13 categories of mobility/diffusibility options further increase the total process variations available.





Clearly, an enormous range of materials-processing alternatives can be indexed by a finite number of decision nodes. One might imagine a very large, complex, but finite extraction machine comprised of 35-40 process categories, each capable of performing an operation described in figures 4.21 or 4.22 (eg, ballistic sublimation, liquid-solid absorption/ion exchange). In addition, each category subsystem is capable of fully monitoring its own input, internal, and output materials streams, and environmental or operating conditions and must have access to detailed knowledge of relevant data and procedures in chemical engineering, physics, and the mathematics necessary to maintain stable operation or to call for help from an overview monitor system. Each processing subsystem communicates extensively with all executive system to select process flows consistent with external factors such as available energy, excess materials, local manufacturability of process components, necessary growth rates and the general environment.

During deployment, the complete package is delivered to a materials source. Representative local raw materials are sampled to select appropriate overall processing options. After selection is made, throughput rates in the process stream are upgraded to full production levels. Output materials are delivered to a generalized manufacturing system which builds larger specialized production units and support systems such as power supplies, mining, and other materials-gathering equipment, transporters, and related items.

In the most general terms, the Materials Processing System reduces variety in the local environment by absorbing unknown or chaotic resources and producing numerous output streams of well characterized industrial materials. Variety reduction is accomplished by definite and finite sequences of analytic operations. The analysis task, though large, is finite. The next step, manufacturing, involves the production of possibly an infinite number of forms, hence will likely require different mathematical and computational approaches.

The concept of a self-contained regenerative processing unit affords an interesting didactic tool. What tasks would be required for the unit to manufacture a collection of locally appropriate processing subsystems? What "cognitive structures" are necessary to organize and to direct the activities of the manufacturing units and the 35-45 analytic cells? Further questions regarding possible tasks include:


 * What physical operations and observations must be conducted in each process category?
 * What equipment types are common to various categories of materials processing, materials transfer, and storage needs?
 * What chemicals are essential for the materials processing capabilities desired?
 * Have any process categories been omitted?
 * What physical knowledge of processing operations must be embedded in directly associated machine intelligence (MI) units?
 * What are the necessary relations between extent of exploration observations, initial test processing, and build-up to large-scale processing?
 * How many process paths should the overall system physically explore? To what extent, and how, should theoretical understanding and limited observations be used to rule out the vast majority of processing alternatives to permit early focus on adequate production sequences?
 * How can new knowledge acquired in operations in new environments and with new compounds be incorporated into the MI system?
 * What principles of overall management must the system obey to ensure survival and growth?
 * What are the fundamental ultimate limits to the ability of self-regenerative systems to convert "as found" resources into industrial feedstock? Are there any essential elements which limit growth by virtue of their limited natural abundance?
 * How can an understanding of physical principles be incorporated into the overall management system to direct operations?

Generalized manufacturing. Figure 4.23 illustrates the generalized manufacturing process. Units 2-8 suggest the flow of formal decisions (along a number of "information transfer loops") and material items which finally result in products. The management unit directs the entire enterprise in response to internal and external opportunities and restrictions. Development of new products requires participation of the entire system, whereas manufacture of repetitive output focuses on providing smooth production flows through units 4-8 guided by management. This schema explicitly refers to the manufacture of "hard products" such as telephones, automobiles, and structural beams, but a generally similar methodology also applies in the preparation Of made-to-order chemical compounds. Thus, the reduced chemical feedstock discussed earlier may supply material to logistics (8) for input to manufacturing processing.



Considerable progress in automation and computer assistance have been made in the functional areas of design (2: computer aided design), parts fabrication (4: computer aided manufacturing), logistics (7: computer aided testing), and management support (1). If extension of state-of-the-art practices is focused on space operations, further advancements readily may be visualized in parts fabrication (4: eg, flexible machining systems), materials handling (5: e.g., automated storage systems and transfer lines, retrieval, parts presentation), assembly (6: e.g., robots with vision and human-like coordination), and inspection and system testing (7: eg., physical examination using vision, sonics, X-rays, or configuration as when checking computer microchip integrity).

Major additional research is necessary In process planning (3), handling (5), assembly (6), and inspection and system testing (7) in order to fully develop autonomous SMF. Although machine intelligence systems are appropriate in all phases of manufacturing, the most advanced applications will be in management, design, and process planning.

There is a fundamental difference between generalized materials processing and manufacturing. In the former (production of "standardized" industrial materials) the system is designed to reduce variety of originally random or unstructured resources. There are a finite number of chemical elements and a finite but extremely large collection of processes and process flows by which chemical elements may be derived from primary native materials. On the other hand, manufacturing processes presumably can impress virtually an infinite range of patterns upon the matter and energy of the Universe. Substitutions of materials and alternate solutions to various engineering challenges are manifestations of the diversity possible. Parts fabrication is the "materials" focus of manufacturing: as shown in figure 4.23, there are four major steps - parts formation, secondary finishing, finishing, and assembling - with matter flowing generally from one stage sequentially to the next.

Table 4.24 by Waldron (Criswell, 1979) presents a non-inclusive functional taxonomy of manufacturing processes which is organized differently from table 4.17. With few exceptions all may be applied to advantage in one or all of the four stages of manufacturing. Each can be used to produce parts of arbitrary size, form, dimensional accuracy, composition, and other collective properties (e.g., magnetic susceptibility, tensional strength, thermal conductivity, switching speeds), so it is clear that a continuously growing diversity of products is possible. Thus, manufacturing intrinsically requires machine intelligence systems to create novel forms embedded in nonterrestrial materials. In turn, these "matter patterns" might be used to control nonmaterial flows of electric and magnetic patterns, momentum, photons and information - the key to further propagation of new pattern production.

The following is a list of research challenges extending from the broadest issues of "matter patterns" to the present state-of-the-art of machine intelligence as applied to design, process planning, and management units depicted in figure 4.23:


 * Creation of world models and methods of identifying "needs" for materials, energy sources, products, etc., which the system must provide for further growth.
 * Observational and communications means and strategies by which world models can be extended, compared to external realities, and then needs recognized and fulfillments confirmed.
 * Computational strategies for optimal uses of the means of production and the resources for creating new products.
 * A method of creating, analyzing, and testing new designs derived from validated theoretical concepts or empirically justified knowledge (i.e., that something works). A similar need exists in the task area of assembly in which knowledge of the desired functions of a device or system can be referred to in the assembly procedure rather than referencing only configurational information or combinatorial blocks in a sequence of assembly steps.
 * Some means of representing the resources of a production system and a formalism for process planning tasks.

The scientific and engineering communities continually strive, in a somewhat uncoordinated manner, to develop new comprehensive physical theories and then apply them to the creation of new material systems. A new scientific/ engineering discipline is needed which explicitly and systematically pursues the following related tasks:


 * Document the historically evolving capability of humanity to impress patterns onto matter, the quality of life as patterning ability becomes more sophisticated, the physical dimensions of pattern impressment, the interaction of new patterns by which even more comprehensive orderings may evolve, and the relationship between physical control over matter-energy and the socially based field of economics.
 * Investigate on very fundamental levels the interrelations among information, entropy, negative entropy, self-organizing systems, and self-reproducing systems. This study should incorporate the latest thinking from the fields of physics, mathematics, and the life sciences in an attempt to create a model or theory of the extent to which regenerative and possibly self-aware designs may be impressed onto local and wider regions of the Universe - a "general theory of matter patterns."
 * Seek the transforms which can be employed at any stage of development to create higher orders of matter patterns.

Human thoughts and conversations typically are conducted using "object"- and "action"-based words learned during childhood. Deeper and more widely applicable symbolic manipulations may be derivable from the mathematical fields of group/set theory, topology, and from the physical and social sciences. A long-term research program should seek to construct a "relationally deep" natural language for human beings and to develop systems for teaching the language both to adults and children. In effect this program would strive to understand intelligence as an entity unto itself and would attempt to explore, identify, and implement more capable "intelligence software" into both life-based and machine-based systems.