Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.
OCR for page 36
Report of the Workshop on Biology-based Technology to Enhance Human Well-being and Function in Extended Space Exploration 3 Enhancing Human Presence and Function in Space Exploration STUDY APPROACH To identify areas where the use of biology-based technology might enhance human function in space exploration, session 2 participants developed a generalized framework outlining how biological concepts and principles relate to various types of functions (see Figure 3.1). The framework provides a structured way of analyzing how biological systems and concepts (the left-hand column) can be integrated into advanced technology (the middle column) to provide a range of specific functions (the right-hand column). In examining the list of biological concepts, mechanisms, and methods of analyzing or modeling biological systems, one can envision how they might be applied to perform different functions. Metabolism is a method for storing and using energy, neurophysiology enables human adaptation to different environments and provides mechanisms for modifying that adaptation, modeling of insect behavior reveals different methods for exploring unknown terrain, and so on. The middle column provides examples of technologies that can implement some biological functions. For instance, very large scale integration (VLSI) and micro-electromechanical systems (MEMS) can integrate multiple functions into large-scale computing and actuation platforms. One such platform is an artificial retina, which can limit the resolution of pixels away from the center of the field of view and perform computations using the incoming data to reduce bandwidth requirements. MEMS-based microrobots have been proposed that can float on a breeze and use integrated power, computing, and actuation to explore unknown terrain. The right-hand column indicates some of the applications of biologically inspired technology. For example, an insect antenna could be used as a model for a biosensor system in which living cells are grown on silicon chips for the purpose of sensing radiation, toxins, or other molecules. Or, the ability to access long-term memory could serve as a computational model for a robot assistant that can
OCR for page 37
Report of the Workshop on Biology-based Technology to Enhance Human Well-being and Function in Extended Space Exploration FIGURE 3.1 Generalized framework outlining how biological concepts and principles relate to various types of functions. suggest modifications to experiments based on previous experience. Session 2 participants used this framework as the basis for a systems approach to identifying how specific biological activities might be exploited in practical ways to enhance human function in space. Relying on input from NASA, the group identified and inferred some of the functions of astronauts that need to be enhanced. Possible biological approaches to meeting those needs were then explored. The proposed solutions emphasized biological principles and analogies as opposed to living or bioengineered systems. The analysis is limited in detail because NASA described its requirements only in broad terms. Session participants attempted to elicit additional details from workshop presenters but still had to make some assumptions about practical needs. Thus, the suggestions provided in this chapter need to be examined further by NASA in the context of defined requirements. A number of general themes were discussed that influenced the analysis of needs and possible solutions. First, functional requirements for space systems differ from those for systems on Earth in a number of ways. Limitations on size and weight are more severe in the space environment, redundancies are needed to prevent system failures on the first error, and higher costs are tolerated in space systems for equivalent benefits. Second, similarities between deep space and the deep ocean—both are extreme environments far removed from home bases—suggest a potential for transferring diving technologies and concepts to the space program. Third, biological principles suggest the merit of processes that are inherently simple and evolutionary, as opposed to complex and excessively mechanical.
OCR for page 38
Report of the Workshop on Biology-based Technology to Enhance Human Well-being and Function in Extended Space Exploration Fourth, there is a need to balance the tasks assigned to machines versus those assigned to humans. For example, machines could perform routine functions and notify the astronaut only if problems arise or decisions need to be made. The space shuttle is fully automated for ascent and landing except for lowering the landing gear. However, 15,000 keystrokes are needed on board for a 10-day mission. Some of these operations might be automated, so as to free astronauts to perform critical mission tasks. Yet some astronauts prefer to be more involved in operations. There is also a possibility that automation can go too far. One leading manufacturer, Toyota, has been reducing automation levels in recent years, having found that properly designed factories using simple, robust processes can enable humans to work very fast. FUNCTIONAL NEEDS Before humans can visit and live on Mars, the functional capabilities of astronauts need to be radically improved and enhanced, beyond even the technologically enhanced capabilities of humans on Earth. In addition, it would be helpful to compensate for the loss of function normally experienced as a result of the constraints of the space environment and equipment. The enhancements are needed in both human presence (i.e., sensory and information processing capabilities) and human functions (i.e., manipulation and locomotion capabilities). For discussion purposes, the functions of astronauts that need to be enhanced were organized into the following categories: perception, manipulation and locomotion, cognition, and systems and computation. These problems and needs dictated the types of promising biology-based research identified by session 2 participants. Perception Astronauts sometimes experience perceptual difficulties. Spatial orientation in microgravity is a particular problem, especially during extravehicular activity (EVA) when there are no gravitational reference points; familiar cues, such as artificial horizons, are needed. Problems have also been reported with hand-eye coordination during long flights and blurriness of the visual field when exiting and reentering 1 G. Problems with depth perception were reported on the Russian space station MIR. Astronauts also may experience difficulties with proprioception (i.e., knowledge of where limbs are) and motor-sensory coordination. Manipulation and Locomotion Posture, locomotion, and balance parameters change when gravity is absent. In microgravity, muscles are loaded and used differently. Muscle mass and cardiovascular capability are lost, and so special exercises are crucial (Desplanches, 1997). Bone can be lost during long missions and may not fully regenerate. Because manipulation is difficult in space, small, lightweight tools that are easy to handle are needed, along with special posture strategies.
OCR for page 39
Report of the Workshop on Biology-based Technology to Enhance Human Well-being and Function in Extended Space Exploration Particular problems are reported with constrained motion in spacesuits, which are pressurized at 4 to 5 pounds per square inch. Fatigue is also common during EVA. Cognition The word “cognition” is often used in computer science-related fields to denote the level of activities that require “understanding” of what is going on, rather than merely signal-level reaction. The primary orientation problem is adaption to microgravity and then readaptation to 1 G when returning to Earth. Symptoms include motion sickness, loss of coordination, and impaired ability to work. There is some concern that astronauts may not function well upon arrival at Mars. Readaptation to 1 G is the more difficult problem. After a shuttle mission, the readaptation period is 4 to 8 days; this time period is proportional to the period of weightlessness and can last just as long. There is a “fragile” period of adaptation when an astronaut can fluctuate between the microgravity and 1-G states. Systems and Computation for Mission Planning and Execution Systems are defined here as encompassing both human and technological elements. System and computation problems are reported with software reliability, control of collective behavior, and human factors, for example. The session participants focused on the mission planning and problem-solving aspects of these problems rather than psychosocial issues and human factors. POTENTIAL BIOLOGY-BASED RESEARCH OPPORTUNITIES Improve Space Suit Design Life support, rather than function and comfort, has been the driver of traditional space suit design. Not surprisingly, difficulties have been reported with fatigue during EVAs and the performance of certain tasks, from mission-related activities to simply scratching an itch. Current-generation space suits need to be redesigned in any case, because they are too heavy for use on Mars and their multilayer insulation would lose most of its effectiveness in the martian atmosphere. As part of the redesign effort, biological concepts could be introduced into space suit design to enhance astronaut function and comfort and provide reliable, mobile, dexterous, comfortable, and easily maintained protection. Humans are extremely flexible, dexterous, and coordinated, but they have limited manipulation and locomotion capabilities when in a space suit, especially when working at extremely small dimensions. Biomechanical concepts could be integrated into suit design to improve task performance. For example, the human wrist has the greatest finger dexterity and hand grip when positioned at a 40-degree angle (see Box 3.1). Current-generation suits do not
OCR for page 40
Report of the Workshop on Biology-based Technology to Enhance Human Well-being and Function in Extended Space Exploration maintain the wrist at that angle. A suit might be designed to have two wrist settings, one for resting and one for tasks requiring dexterity or grip. Current suits are passive, in that they do not assist with movement. With an active compliance control approach, it should be possible to design a suit that could support the astronaut's activities. The suits could be designed to be active through the use of a wide range of actuators combined with MEMS technology for sensing intended motions, a capability that has advanced considerably in recent years (Epstein and Senturia, 1997). Actuators are already used in prosthetic devices to provide guidance. The suit could also be designed to diagnose and repair its own malfunctions. External sensors could be incorporated into a suit to produce haptic and other sensory feedback to the astronaut. For example, electrical stimulation with stochastic (random) frequency could be used to improve the astronaut's sense of touch. Another possibility is to use low-amplitude muscle vibrators in which the frequency is modulated stochastically. Sensors could be part of a “smart glove” system. An information filter would be needed to ensure that the astronaut is not overloaded with irrelevant sensory data. Box 3.1 Human Biomechanics and Space Suit Design Human biomechanics is complex and needs to be considered in the development of human-machine interfaces such as in space suit design. A simple example is the production of maximum finger grip. The muscle tendons that cause the fingers to flex run from muscles in the forearm across multiple joints, including the wrist, to end on the bones of the fingers. The ability of the muscle to generate force is very dependent on the length of the muscle. Different positions of the wrist influence the length of the muscle and the amount of force that can be generated. If the wrist is palmar-flexed, then the muscle is shortened and cannot generate as much force. The optimal wrist position for maximal finger grip is approximately 40 degrees of wrist dorsiflexion. If a space suit glove constrains the wrist position to some other degree of flexion, muscle force will not be optimal, resulting in reduced dexterity and increased fatigue. Similar considerations extend to almost every movement because nearly every muscle and tendon extends over multiple joints. Human joints are complicated, unlike simple hinge mechanical joints. The axis of rotation is not stationary but may move in a complicated pattern. Therefore, space suits and tools need to be developed to allow additional physiological movement of the limbs for maximal efficiency. Again, the wrist is a good example. As the wrist bends, the actual length of the arm and hand lengthens or shortens because the wrist glides forward or backward. This changes the mechanical lever for movements and therefore could influence performance. Although this may not be an issue in the normal Earth environment, the highly demanding environment of space and Mars may exact a severe penalty if human motor performance is constrained to a level that is not physiologically optimal. It may be possible to exploit the complexities of human biomechanics. The increased complexity of human movement enables the control of greater degrees of freedom, which can be translated into expanded dimensions of information or process control. Thus, tapping into the greater complexities of human movement could increase the amount and complexity of information communicated across the human-machine interface. A suit might also be designed to provide spatial orientation cues in microgravity. Research aimed at providing these cues through galvanic stimulation (electrical or magnetic) has been under way for many years. Some success has been achieved with vestibular prostheses, such as the “Cuban boot,” in which an expandable bladder applies pressure to the bottom of the foot, and the “Pensacola vest,” in which vibrators are embedded to provide cues as orientation
OCR for page 41
Report of the Workshop on Biology-based Technology to Enhance Human Well-being and Function in Extended Space Exploration changes. Visual cues can also help with spatial orientation (Kornilova, 1997). (See section below on visual input/output.) Materials are an essential issue in space suit design. Biological materials that could provide useful concepts for space suits include the exoskeletal structures of insects, which are quite dexterous, and sharkskin, which has a cross-hatched fiber orientation that provides both flexibility and structural integrity (Wainright et al., 1978). There are many examples of cross-spiral-wrapped biological structures with well-studied and highly desirable properties such as axial bending (e.g., endomysial connective tissue, worm bodies, giraffe fascia). Cross-hatched material might be used in a lightweight “space leotard ” similar to a skin-diving wetsuit. The leotard would counter pressure with elasticity but be transparent to sweat. It could be combined with a helmet and worn under loose coveralls, with bladders in body cavities. Improved space suit technology could be the topic of a follow-on workshop. This research is short term and could bear fruit within 5 to 10 years. Maintain and Improve Physical State A fully effective exercise regimen has yet to be designed for maintaining or restoring the cardiovascular and musculoskeletal function of astronauts (Convertino, 1996; Desplanches, 1997). Biological principles might be applied to exercise concepts to help maintain the astronaut's physical state, including muscle and cardiovascular strength and hand-eye coordination.1 The exercise concept mimics a form of resistance training. Gymnastics has been deemphasized by NASA recently in favor of mechanical exercise devices, but some of the best exercise for astronauts involves jumping across the spacecraft and doing flip turns (as is often done in lap swimming) off the walls, because launching off the wall provides some 1 G-like acceleration. The biologically inspired twist on this idea involves having the astronaut exercise in a large sac (emulating an embryonic sac) or “bungee suit” with elastic properties. It might also be useful to design a “microgravity gymnasium” where astronauts could participate in gymnastics. The embryonic sac analog and the microgravity gymnasium might prove to be more effective than, or a valuable supplement to, the current generation of NASA exercise machines (primarily the treadmill). The bungee bag could also be used for sleeping and storage. The exercise regimen could be combined with virtual reality and computer games using joysticks of different sizes and configuration to maintain hand-eye coordination. Appropriate integration of tasks may serve to make exercise more enjoyable. These projects could be informed by the recent development of rehabilitation strategies for persons with long-standing motor disorders or whose muscles have undergone a prolonged period of disuse. This work demonstrates the importance of the pragmatic (i.e., function- 1 Basic biological research might also be helpful. Controlled studies of natural and engineered tissues (e.g., using bioreactors under the conditions of simulated and actual microgravity) could help scientists understand the effects of spaceflight on human tissues and develop and test potential countermeasures (e.g., drugs).
OCR for page 42
Report of the Workshop on Biology-based Technology to Enhance Human Well-being and Function in Extended Space Exploration oriented) approach, describes attempts at using virtual reality in motor rehabilitation, and details adaptive changes within the central nervous system (Davis and Burton, 1991; Latash and Anson, 1996). The possibility of the existence of direct human sensing under local conditions for medical diagnosis is being investigated at major medical centers, including the University of Virginia, University of Colorado, Columbia University, Duke University (Meredith, 1997), and Harvard University (Roush, 1997). The Office of Alternative Medicine at the National Institutes of Health has awarded several grants to investigate these abilities in controlled studies. Given the limited medical resources that will be available in long-term spaceflight or colonization, and the remoteness of the astronauts from medical facilities, NASA could monitor research on alternative medicine approaches to see if they can be applied in a space medicine context. The exercise research projects are short term. Enhance Adaptation to New Environments NASA has experimented with various methods (e.g., biofeedback) for accelerating astronaut adaptation to microgravity and 1-G environments but has yet to find an optimal solution. A biologically inspired approach to accelerating the adaptation period might be found in understanding and manipulating the fragile state of transition between microgravity and 1 G. It might be possible to influence adaptation through pharmacological approaches that increase the plasticity of the brain. There is evidence from both everyday life and biomedical research that points to an inherent biological capability for “dual adaptation ” (Welch et al., 1993). These observations suggest that humans can adapt to a dual-mode existence. Examples include the common experience with wearing glasses: It takes a while to see comfortably out of a new pair, but, once the body's systems adapt, the eyes can automatically focus with or without the glasses (the experience with bifocals is similar). Another example is task-specific dystonia, in which a person experiences hand cramps, for example, when performing certain tasks but not others. In other words, the biological response differs from mode to mode (Koller, 1989; Byl et al., 1996). Researchers have found evidence of a dual-mode capability in the brain (Merzenich and Jenkins, 1996). It may be possible for the brain to operate in multiple modes with the ability to switch from one to the other. This capacity could be explored so as to train the nervous system to operate in two modes: one appropriate to microgravity and the other to 1-G environments. The subject would then be able to switch rapidly from one mode to the other. There is evidence to suggest multiple representations in the brain; an example is the Necker Cube illusion. The question is whether a particular face of the cube is in front or in back. In fact, the perception of one orientation can alternate with perception of the other. Another example is the illusion of two faces facing each other. Again, the illusion can be perceived as one or the other but not both, and the perception can alternate. There is another example that is the result of training (Martin et al., 1996). Individuals were trained to throw darts while wearing a pair of eyeglass prisms. These prisms shifted the visual world in one direction. At first, the subject would err in throwing the darts to the side of
OCR for page 43
Report of the Workshop on Biology-based Technology to Enhance Human Well-being and Function in Extended Space Exploration the target in the direction in which the image was shifted. The subject then learned to compensate and consistently reach the target. When the subject took the glasses off, the subject would err in the opposite direction. However, when the subject used the other hand to throw the dart, the subject was accurate. Thus, training to adapt to the novel environment affected the use of one arm but not the other. Also, the subject could instantly use the untrained arm in the usual environment while the trained arm would err. Thus, there were two states: one capable of functioning in the normal visual environment and the second adapted to the novel environment. These two states also coexisted. Extensive research is under way on how information is remembered at the cellular and molecular levels. Adaption to the space environment or readaption to the Earth environment may represent a form of learning. Neuroscience has discovered a great deal about the biochemical basis of learning, thus leading to the possibility of using medications to affect learning and, hence, adaption. In the future, medication may help to shorten the time needed for readaption. Already there are examples indicating that the combination of pharmacological agents and physical training can facilitate motor learning (Walker-Gatson et al., 1995). Although current studies used amphetamine-like compounds, it is highly likely that other safer and more effective compounds can be developed. A specific group of receptors have been identified as involved in learning. They are known as N-methyl-D-aspartate or NMDA receptors (Blanchet et al., 1997). Research indicates that activation of these receptors combined with electrophysiological activity (as could accompany training) can lead to changes in nerve cell structure and function consistent with learning (McNaughton et al., 1994). These developments are part of the rapid increase in understanding of the central nervous system and its plasticity. A combination of pharmacological intervention and appropriate training and exercise might effectively prepare astronauts for dual adaptation. It might also be possible to target specific functions and thus counteract some of the more critical adaptability problems. This is a long-term research project. However, rapid advances in understanding of the biological basis of learning and adaptation may enable relevant results to be achieved in the near term (5 years). Enhance Visual Input/Output Visual capabilities play an important role in the human exploration of space, not only in documenting the environment but also in maintaining spatial orientation. Human vision has a number of limitations, including poor three-dimensional measurement capabilities. Technological systems offer opportunities for enhancing human vision and providing remote telepresence (e.g., over-the-horizon sight). However, artificial sensing devices currently tend to be bulky, primitive, and in most cases far less sensitive, precise, or adaptive than their human or biological counterparts. Biological models and concepts might be applied to enhance sensing capabilities or even provide new sensory modalities (e.g., nonvisible or nonaudible domains, or
OCR for page 44
Report of the Workshop on Biology-based Technology to Enhance Human Well-being and Function in Extended Space Exploration over-the-horizon vision). Biological principles could also be applied to enhance astronauts' ability to maintain spatial orientation in microgravity. Visual technologies are under active investigation in many sectors, including military research laboratories, the medical community, the entertainment industry, and academia.2 Considerable hardware exists, but the state of the art needs to be improved. Furthermore, it is not clear which types of visual devices would be most useful to NASA, or what information would have to be displayed. Current-generation head-mounted displays are bulky and typically offer a restricted (e.g., 90-degree) field of view. Common complaints about heads-up displays in aviation include data overload and too much nongeographical (i.e., text) information. Attempts to increase display resolution result in bandwidth limitations. Another approach to implementing head-mounted displays involves writing directly on the retina with a laser (Tidwell et al., 1995). Extensive research is under way to design an artificial retina consisting of a sensor and processor on a single chip; the challenge is to reduce data and bandwidth needs and increase dynamic range. The artificial retina is an example of computational sensors, which combine computation and sensing to increase performance and capabilities over those enabled by standard sensing and computing modalities.3 These sensors integrate VLSI elements, either analog or digital, directly with sensing elements or nearby in a tightly coupled, on-chip manner, or use geometry or material properties to achieve computational gains. Computation is typically performed as a local operation with a distributed model of the sensory data, which, for example, can take advantage of the parallel nature of computing with a two-dimensional image plane (Kanade and Bajcsy, 1993). Efforts are being made to miniaturize head-mounted displays and to design augmented, hybrid reality systems consisting of image overlays (Azuma, 1997). The idea is to replace the goggles with a camera and project the camera's field of view onto the eye. Such a system might help an astronaut see, with the help of exploratory robots or other sensing systems, a projection of what lies over the next hill. Another possible solution is a programmable lens device, a biology-based analog to the flexible cornea, which focuses by changing shape. A programmable lens could, for example, improve vision in poor lighting by picking out and enhancing important features. A similar system currently being worked on involves binary optics, where lenses are built up directly on the chip using standard VLSI and ultra-large-scale-integration techniques (Stern, 1996). Both of these technologies would be very important for decreasing system size and mass and increasing performance for use in space systems. Many years ago, a device was invented that projects visual images from a helmet-mounted camera onto a person's back or abdomen, using a vibrating-pin matrix similar to a 2 Some of the current activity in this area is described on the World Wide Web site of Carnegie Mellon University's computer vision home page (http://www.cs.cmu.edu/afs/cs/project/cil/www/vision.html). 3 One of the most important aspects of biological sensors is their heavy reliance on efferent systems (e.g., oculomotor steering, outer hair cell motility in the cochlea, fusimotor control of muscle spindles) that are much more sophisticated than the usual low-level functions of electronic sensors. Many of these functions could be incorporated into existing electromechanical technology once their principles of operation are understood.
OCR for page 45
Report of the Workshop on Biology-based Technology to Enhance Human Well-being and Function in Extended Space Exploration Braille reader. This device was used with blind subjects, who gained the ability to read and, once they had acquired video-to-hand coordination, to manipulate objects (Bach-y-Rita et al., 1969). Such an instrument might be implemented in an elegant fashion using modern technology. An astronaut with normal vision might gain some additional capabilities, especially if the camera could sense, for example, infrared or ultraviolet light. Like other artificial sensing systems, this technology would need to be vastly improved to achieve its theoretical potential. Another important aspect of vision is its role in maintaining spatial orientation (Matin and Fox, 1989; Matin and Li, 1995). Living things respond to primeval visual stimuli. Pilots can attest to the importance of having a visual horizon; indeed, the need for this cue is so significant that birds have horizon-detector ganglion cells in their retinas (Maturana and Frenk, 1963). A visual horizon might be created on the interior of the transit vehicle, with different colors visible above and below a clearly demarcated line. Astronauts would become accustomed to seeing this visual cue in a spacecraft on the ground and would develop an internal reference system. For further realism, the lighting could mimic the typical human experience, in which light enters the retina from above. “Uplighting,” by contrast, appears unusual, even sinister. Accordingly, the upper horizon within the spacecraft could be illuminated with white light, and the lower horizon with blue light, as it appears to a person diving in the sea. All these research efforts would achieve results in the short term. Develop Synergistic Human-Robot Systems Robots perform many useful functions in space exploration. Their role may be most obvious on unmanned missions, when they serve as human surrogates, but robots can also enhance the presence and function of astronauts by providing various types of assistance. There may be opportunities to apply biological principles in robot development to help meet NASA's needs for effective human-machine collaboration, improved situational awareness, and optimal decision making. For example, the development of software that senses its own and an astronaut's anxiety could provide robots with the “brains” needed to extend mission capabilities while also enhancing the reliability and configurability of the software. This idea builds on the research community's long-standing use of the biological knowledge base as a source of ideas for the design of algorithms and computational mechanisms for data Box 3.2 Collaborative Multirobot Systems Collaborative multirobot systems draw on the concept of large numbers of agents performing similar tasks, as often seen in the insect kingdom. Recent advances in MEMS, microactuators, and high-density power sources have made possible the mass production of a large number of small, complex devices that can sense and affect their environment. A system of several small, cooperating robots provides advantages such as graceful degradation in case of the loss of individual robots, robust inter-communication among robots and with their human counterparts, reconfigurability to attain varying objectives, and the ability to deploy individual specialized robots. This type of system could be used to enhance astronauts' capabilities to do large-scale mapping, detailed exploration of regions of interest to build three-dimensional maps that astronauts could explore through virtual reality to determine regions of interest (similar to Sojourner), and automated sampling of rocks and soil.
OCR for page 46
Report of the Workshop on Biology-based Technology to Enhance Human Well-being and Function in Extended Space Exploration processing (neural networks and DNA computing are examples). Other biologically inspired robot systems are conceivable (an example is provided in Box 3.2). Robots are the focus of considerable academic research4 but their capabilities remain primitive in comparison to those of biological organisms. For example, in the last 5 years, mobile research robots have just begun to succeed in tasks such as navigation of roads and hallways. These successes follow from the increase in robot computer power from 1 million instructions per second (MIPS) to several hundred MIPS, which corresponds to a nervous system less powerful than that of an insect (Moravec, 1998). One way to enhance robot capabilities might be to develop controller software that senses anxiety. The importance of emotional states is demonstrated by their strong effect on human decision making. There is evidence that individuals whose emotional centers have been damaged by disease make poor decisions (Kandel et al., 1991; Pinel, 1993; Goleman, 1995; Greenspan and Benderly, 1997). The biological principles at work are neuronal. Human behavior and problem solving emerge from resource competition and collaboration, and the environment of the competition is strongly affected by emotion. Furthermore, the learning process is mediated and reinforced by emotion: Humans generally repeat things that feel good and avoid things that do not. Humans also remember events and skills associated with the current emotional state. Based on this biological model, software for robotics systems could be designed to exhibit evolving system behavior mediated by emotion and anxiety. Modules would compete to perform different functions, and the learning process would be augmented by emotion. A robot of this type would be especially useful as a component of a synergistic human-robot system. The robot would be attuned to both its own and the astronaut's feelings and provide backup (i.e., system robustness) against the introduction of new elements and situations. Communication between the astronaut and robot concerning their respective emotional states could help reinforce the positive or negative results of the associated actions and thereby promote learning. A sophisticated robot could even take over the astronaut's tasks in times of stress. A physiological response (e.g., anxiety) could trigger the software to take on certain functions for the human and delay nonessential activities. Interfaces could be designed to enable human comprehension of system data without information overload, and to communicate human intentions to the robot. There are interesting human-computer interface possibilities, perhaps using modes of communication involving electrical fields and currents or visual information. Verbal communication is perhaps the most natural medium for humans, and the ability of computer systems to communicate verbally is generally appreciated. However, human verbal communication is highly complex and goes well beyond literal syntax and semantics. There are paralinguistic elements to be considered such as intonation, rate of speech, level of formality, turn-taking, and verbal etiquette (personal 4 Some of the research activity in this area is described on the World Wide Web site of Carnegie Mellon University's Robotics Institute (http://www.ri.cmu.edu) and Microdynamics Systems Laboratory (www.cs.cmu.edu/~msl/msl_hyperlinks.html).
OCR for page 47
Report of the Workshop on Biology-based Technology to Enhance Human Well-being and Function in Extended Space Exploration communication to Ewin Montgomery by Lyn Turkstrata, Department of Communication, Case Western Reserve University).5 Apart from their value in mission execution, effective human-computer interactions might also help ease the psychological demands of prolonged spaceflight. For example, a robot attuned to anxiety and the paralinguistic elements of communication might help identify and deal with situations in which an astronaut needs to “psychologically escape” from other crew members. There are already examples (see the section on PI-in-a-box below) of computers programmed to generate appropriate and emphatic responses to human input. The design of software for telerobotics systems incorporating emotion-mediated behavior and learning is a long-term research project. Monitor Cognitive States The influence of emotion on decision making (see previous section) suggests there may be some value in monitoring astronauts' cognitive states. For example, it would be useful to have a process for identifying and dealing with anxiety, which can be correlated with negative performance. It may not be feasible to teach astronauts to be self-aware and express their feelings, because there appears to be a cultural bias against communicating any problems, especially given the public nature of shuttle broadcasts. Anecdotal reports suggest that astronauts' inhibitions against revealing their emotions have compromised abilities to accomplish tasks in the past. Physiological monitoring of cognitive and emotional states could provide confidential biofeedback to promote relaxation. There is already a large literature and clinical practice dealing with these techniques (Basmajian, 1989). The ability to monitor brain states could have additional benefits. The results of monitoring could be displayed solely to the astronauts and ground-support personnel. Physiological variables can be monitored noninvasively using various techniques. A traditional means of gauging emotions or arousal is the lie detector, which measures blood pressure, respiration, and galvanic skin response. Other standard methods (e.g., electroencephalography [EEG] and electromyography [EMG]) are well established but often awkward, because of the need to attach electrodes to the subject, who must then remain still. Implanted sensors and telemetry can also be used. It is also possible to monitor outward signs of emotion with software. A system is being developed for facial-expression recognition that could be modified to look for signs of emotional state (Cohn et al., 1997). Magnetoencephalography (MEG) may offer some advantages over EEG because it is less affected by tissue conductivity (Haueisen et al., 1997). Therefore, MEG may be more accurate in 5 The complexity of paralinguistic elements provides many dimensions for encoding information. Indeed, humans are adept at perceiving and operating on the basis of these paralinguistic elements. For example, factual information could be conveyed in the syntax and semantics while the priority of the information is encoded in the paralinguistic elements. This approach would allow the human to process the factual information and the priority simultaneously and relieve the computer of having to convey the information sequentially, perhaps offering a significant advantage in time-critical operations.
OCR for page 48
Report of the Workshop on Biology-based Technology to Enhance Human Well-being and Function in Extended Space Exploration localizing some electrical sources, especially those deep in the brain (Stok, 1987; Hamalainen and Sarvas, 1989; Lewine and Orrison, 1995), potentially accessing relevant neuronal events for the purpose of monitoring or even short-circuiting them. MEG also provides almost instantaneous feedback, making it attractive for cognitive monitoring purposes. As with EEG, MEG responses can be correlated with measures of cognition that are affected by stress and other variables (Rogers, 1994). Compared with EEG, the technology for recording MEG is far more expensive but much simpler to use, because it is not necessary to spend time carefully attaching large arrays of electrodes. MEG is based on the use of superconducting quantum interference devices (SQUIDS) to detect very small magnetic fields. A SQUID cryogenic cap or helmet for recording brain waves, and perhaps providing immediate feedback to the astronaut using a small display, is an intriguing concept. The technology exists to make very small devices in which the coils are flat and require minimal energy. Indeed, the SQUID cap concept may be particularly appropriate in the space environment, where the temperatures are cold enough, theoretically, to make the SQUIDS superconducting. Even if this theory does not hold up, progress in refrigeration for high-temperature SQUIDS operating at liquid-nitrogen temperatures might make the idea feasible. The Cryogenics Group at NASA Goddard Space Center is developing small Sterling cryo-coolers in conjunction with Lockheed that can probably be miniaturized within 10 years to the size of a 1-kg helmet containing a large array of high-temperature SQUID sensors (Stephen Castles, NASA Goddard Space Center, January 1998, personal communication). Pulse-tube coolers may perform as well in a small helmet (Ray Radebough, National Institute of Standards and Technology, January 1998, personal communication), and reverse-Brayton cryo-coolers may have similar performance capabilities (McCormick et al., 1997; Stacy et al., 1997; Radebough, 1997). Another rapidly developing technology is magnetic resonance imaging (MRI). Great strides are being made in developing and improving new applications for MRI, particularly as functional information is provided to accompany the anatomical pictures, which continue to improve in quality and spatial resolution. The combined structural and functional imaging capabilities of MRI offer extraordinary opportunities to determine the parts of the brain associated with specific abilities (Orrison et al., 1995). However, the response of functional MRI (on the order of 1 to 2 seconds, because of the inherent physiology of the blood flow response) is far slower than that of EEG and MEG (on the order of milliseconds), meaning that dynamic cognitive states cannot be monitored effectively with MRI. Furthermore, size, power, and extraneous magnetic field considerations currently prevent the use of MRI in space. EMG is another potentially useful technology, because it can be used to monitor muscle activity for purposes of biofeedback to reduce stress and anxiety (Basmajian, 1989). Most approaches to monitoring cognitive states would require long-term research. The cryogenic SQUID cap might be practical within 5 years and could be the subject of a follow-on workshop.
OCR for page 49
Report of the Workshop on Biology-based Technology to Enhance Human Well-being and Function in Extended Space Exploration Provide PI-in-a-Box The complexity and new challenges associated with missions to Mars necessarily call for dynamic mission planning and execution strategies and improved problem solving. Astronauts might benefit from having instant access to a database of the accumulated experience of previous astronauts. The database could be self-organizing to respond to immediate needs. Algorithms showing relationships between “faults” and related systems could suggest problem-solving options. This concept is variously described as experience on demand, just-in-time training, or PI-in-a-box (Young et al., 1989; Hazelton et al., 1993; Young, 1994). Research is needed concerning how best to organize and present the data. Biology-based concepts could be applied to data presentation. For example, algorithms based on the survival instinct would present data on the most-life-threatening situations first. NASA has performed research on the robotic extension of human presence using concepts such as the “third hand.” A PI-in-a-box might not be able to make decisions but it could provide information and serve as a “buddy” for the astronaut. A companion robot, similar to a deep-sea diving partner or the sled pulled by Antarctic explorers, could contain support materials such as positioning systems, tents for surviving sand storms, additional oxygen for security, and shovels and other tools for exploration. The robot could also provide mental support when the astronaut is challenged. The robot could contain a computer, database, and logistic backup system that would offer a feeling of security. The U.S. military 's tactical information assistants are based on a similar concept. 6 The research needed to develop PI-in-a-box for space applications is long term. SUMMARY The session 2 discussion generated a number of biological concepts and possible applications that may be of use to NASA in extending human presence and function in space exploration. Two topics seem promising enough in the short term to be the subjects of follow-on workshops. One such topic is the design of comfortable, functional spacesuits that incorporate biological concepts, such as 40-degree-angle wrist settings that provide maximum dexterity and grip, and biomolecular materials such as those modeled on strong-yet-dexterous sharkskin. A second workshop could explore the possibility of designing a SQUID cap or helmet that would apply MEG technology to record astronauts ' brain waves and provide feedback on cognitive states. Other R&D areas that might generate practical solutions to NASA's needs in the short term include visual systems and exercises combining gymnastics and virtual reality. Over the long term, biological concepts might be applied to foster the dual adaptation of astronauts to both 1-G and microgravity environments, design synergistic human-robot systems that exhibit 6 Information about tactical information assistants can be found on various World Wide Web sites (http://www.spectrum.ieee.org/publicaccess/1195inf. html).
OCR for page 50
Report of the Workshop on Biology-based Technology to Enhance Human Well-being and Function in Extended Space Exploration emotion-mediated decision making and learning, and develop computational “companion” robots that provide astronauts with rapid access to the combined experience and expertise of their predecessors and enhance both function and a feeling of security. In pursuing any of this work, NASA would benefit from monitoring cutting-edge developments in biology as well as in various technologies that mimic, implement, or exploit biological concepts and capabilities. Technologies of interest include task-specific robots, visual systems, MEMS, and MEG. REFERENCES Azuma, R.T. 1997. A survey of augmented reality. Presence: Teleoperators and Virtual Environments 6(4):355-385. Bach-y-Rita, P., C.C. Collins, F.A. Saunders, B. White, and L. Scadden. 1969. Vision substitution by tactile image projection. Nature 221:963-964. Basmajian, J.W., ed. 1989. Biofeedback: Principles and Practice for Clinicians. Baltimore: Williams and Wilkins. Blanchet, P.J., S.M. Papa, L.V. Metman, M.M. Mouradian, and T.N. Chase. 1997. Modulation of levodopa-induced motor response complications by NMDA antagonists in Parkinson's disease. Neurosci. Biobehav. Rev. 21(4):447-453. Byl, N.N., M.M. Merzenich, and W.M. Jenkins. 1996. A primate genesis model of focal dystonia and repetitive strain injury: Learning-induced dedifferentiation of the representation of the hand in the primary somatosensory cortex in adult monkeys. Neurology 47:508-520. Cohn, J.F., A. Zlochower, J. Lien, Y.T. Wu, and T. Kanade. 1997. Automated face coding: A computer-vision based method of facial expression analysis. Pp. 329-333 in 7th European Conference on Facial Expression, Measurement, and Meaning , N.H. Frijda, ed. Saltzburg, Austria. Convertino, V.A. 1996. Exercise as a countermeasure for physiological adaptation to prolonged spaceflight. Med. Sci. Sports Exercise 288(8):999-1014. Davis, W.E., and A.W. Burton. 1991. Ecological task analysis: Translating movement behavior theory into practice. Adapted Physical Activity Quarterly 8:154-177. Desplanches, D. 1997. Structural and functional adaptations of skeletal muscle to weightlessness . Int. J. Sports Med. 18(Suppl. 4):S259-S264. Epstein, A.H., and S.D. Senturia. 1997. Macro power from micro machinery. Science 26:1211. Goleman, D. 1995. Emotional intelligence: Why It Matters More Than IQ. New York: Bantam Books. Greenspan, S.I., and B.L. Benderly. 1997. The Growth of the Mind and the Endangered Origins of Intelligence . Reading, Mass.: Addison-Wesley.
OCR for page 51
Report of the Workshop on Biology-based Technology to Enhance Human Well-being and Function in Extended Space Exploration Hamalainen, M.S., and J. Sarvas. 1989. Realistic conductivity geometry model of the human head for interpretation of neuromagnetic data. IEEE Trans. Biomed. Eng. 36:165-171. Haueisen J., C. Ramon, M. Eiselt, and H. Brauer. 1997. Influence of tissue resistivities on neuromagnetic fields and electric potential studied with a finite element model of the head. IEEE Trans. Biomed. Eng. 44(8):727-735. Hazelton, L.R., N. Groleau, R.J. Franier, M.M. Compton, S.P. Colombano, and P. Szolovits. 1993. PI in the sky: The astronaut science advisor on SLS-2. Paper presented at the Sixth Annual Space Operations and Research Conference, Houston, Texas. http://www.cs.cmu/edu/afs/cs/project/cil/www/vision.html http://www.cs.cmu.edu/~msl/msl_hyperlinks.html http://www.ri.cmu.edu http://www.spectrum.ieee.org/publicaccess/1195inf.html Kanade, T., and R. Bajcsy. 1993. Computational sensors: A report from the DARPA workshop. Pp. 335-350 in Image Understanding Workshop. Defense Advanced Research Projects Agency, Software and Intelligent Systems Technology Office. San Mateo, Calif.: Morgan Kaufmann Publishers, Inc. Kandel, E.R., J.H. Schwartz, and T.M. Jessell. 1991. Principles of Neural Science, 3rd Ed. Norwalk, Conn.: Appleton & Lange. Koller, W.C. 1989. Task-specific dystonia. Neurology 39(3):457. Kornilova, L.N. 1997. Vestibular function and sensory interaction in altered gravity. Adv. Space Biol. Med. 6:275-313. Latash, M.L., and J.G. Anson. 1996. What are normal movements in atypical populations? Behav. Brain Sci. 19:55-106. Lewine, J.D., and W.W. Orrison. 1995. Magnetoencephalography and magnetic source imaging. Pp. 369-417 in Functional Brain Imaging, W.W. Orrison, J.D. Lewine, J.A. Sanders, and M.F. Hartshorne, eds. St. Louis: Mosby. Martin, T.A., J.G. Keating, H.P. Goodkin, A.J. Bastian, and W.T. Thach. 1996. Throwing while looking through prisms. II. Specificity and storage of multiple gaze-throw calibrations. Brain 119:1199-1211. Matin, L., and C.R. Fox. 1989. Visually perceived eye level and perceived elevation of objects: Linearly additive influence from visual field pitch and from gravity . Vision Res. 29:315-324. Matin, L., and W. Li. 1995. Multimodal basis for egocentric spatial localization and orientation . J. Vest. Res. 6:499-518. Maturana, H.R., and S. Frenk. 1963. Directional movement and horizontal edge detectors in the pigeon retina. Science 142:977-979. McCormick, J.A., W.L. Swift, and H. Sixmith. 1997. Progress on the development of miniature turbomachines for low-capacity reverse-Brayton cryocoolers. Cryocoolers 9, R.G. Ross, Jr., ed. New York: Plenum Press.
OCR for page 52
Report of the Workshop on Biology-based Technology to Enhance Human Well-being and Function in Extended Space Exploration McNaughton, B.L., J. Shen, G. Rao, T.C. Foster, and C.A. Barnes. 1994. Persistent increase of hippocampal presynaptic axon excitability after repetitive electrical simulation: Dependence on N-methyl-D-aspartate receptor activity, nitric-oxide synthase and temperature. Proc. Natl. Acad. Sci. 91:4830-4834. Meredith, D. 1997. It's all in your head. Duke Magazine (Jan.-Feb):16-19. Merzenich, M.M., and W.M. Jenkins. 1996. A primate genesis model of focal dystonia and repetitive strain injury: Learning-induced dedifferentiation of the representation of the hand in the primary somatosensory cortex in adult monkeys. Neurology 47(2):508-520. Moravec, H. 1998. Universal Robots: Object to Person to Spirit. New York: Oxford University Press. Orrison, W.W., J.D. Lewine, J.A. Sanders, and M.F. Hartshore, eds. 1995. Functional Brain Imaging. St. Louis: Mosby. Pinel, J.P.J. 1993. The biopsychology of emotions and mental illness. Chapter 2 in Biopsychology, 2nd Ed. Boston: Allyn & Bacon. Radebough, R. 1997. Advances in cryo coolers. P. 33 in Proceedings of the 16th International Cryogenic Engineering Conference and International Cryogenic Materials Conference. Oxford, England: Elsevier Science. Rogers, R. 1994. Magnetoencephalographic imaging of cognitive processes. Pp. 289-297 in Functional Neuroimaging—Technical Foundations, R.W. Thatcher, M. Hallett, T. Zeffiro, E.R. John, and M. Huerta, eds. New York: Academic Press. Roush, W. 1997. Herbert Benson: Mind-body maverick pushes the envelope. Science 276:357-359. Stern, M.B. 1996. Binary optics: A VLSI-based microoptics technology. Microelectron. Eng. 32(1-4):369-388. Stacy, W.D., T. Pilson, A. Gilbert, and J. Bruning. 1997. Development and demonstration of the Creare 65K standard spacecraft cryocooler. Cryocoolers 9, R.G. Ross, Jr., ed. New York: Plenum Press. Stok, C.J. 1987. The influence of model parameters on EEG/MEG single dipole source estimation. IEEE Trans. Biomed. Eng. 34(4):289-296. Tidwell, M., R.S. Johnston, D. Melville, and T.A. Furness III. 1995. The virtual retinal display—A retinal scanning imaging system. Pp. 325-334 in Proceedings of Virtual Reality World '95. Munich, Germany: IDG Conferences and Seminars. Wainright, S.A., F. Vosburgh, and J.H. Hebrank. 1978. Shark skin: Function in locomotion. Science 202(4369):747-749. Walker-Gatson, E., P. Smith, P. Curtis, H. Unwin, E. Allen, M. Wood, and R. Greenlee. 1995. Amphetamine paired with physical therapy accelerates motor recovery following stroke: Further evidence. Stroke 26:2254-2259. Welch, R.B., B. Bridgeman, S. Anand, and K.E. Browman. 1993. Alternating prism exposure causes dual adaptation and generalization to a novel displacement. Percept. Psychophys. 54:195-204. Young, L.R., S.P. Columbano, G. Haymann-Haber, N. Groleau, P. Szolovits, and D. Rosenthal. 1989. An expert system to advise astronauts during experiments. Proceedings of the International Astronautical Congress, Malaga, Spain.
OCR for page 53
Report of the Workshop on Biology-based Technology to Enhance Human Well-being and Function in Extended Space Exploration Young, L.R. 1994. PI-in-a-box. J. Soc. Instru. Contr. Eng. 33(2):119-122.
Representative terms from entire chapter: