o Abstract - Spatial disorientation (SO) in flight

Transcription

1 Journal of Vestibular Research, Vol. 2, pp , 1992 Printed in the USA. All rights reserved /92 $ Pergamon Press Ltd. THE SPATIAL DISORIENTATION PROBLEM IN THE UNITED STATES AIR FORCE Kent K. Gillingham, MD, PhD Flight Motion Effects Branch, Crew Technology Division, Crew Systems Directorate, Armstrong Laboratory, Brooks Air Force Base, Texas Reprint address: Kent K. Gillingham, MD, PhD, ALlCFTF, Brooks Air Force Base, TX o Abstract - Spatial disorientation (SO) in flight w~tes hundreds of millions of dollars worth of defense capability annually and continues to. kill aircrew. SO results primarily from inadequacies of human visual.and.vestibular sensory systems in the flying environment; but other factors, such as task saturation and distraction, precipitate it. The United States Air Force is conducting a threepronged research and development effort to solve the SO problem. We are attempting 1) to elucidate further the mechanisms of visual and vestibular orientation and disorientation, 2) to develop ground-based and inflight training methods for demonstrating to pilots the potential for SO and the means of coping with it, and 3) to conceive and evaluate new ways to display flight control and performance information so that pilots can maintain accurate spatial orientation. o Keywords - spatial disorientation; disorientation; spatial orientation; pilot vertigo. Definitions An orientational percept is a sense of one's position and motion relative to the plane of the earth's surface. It can be primary (that is, natural), meaning that it is based on ambient visual, vestibular, or other sensations that normally contribute to our orientation in our natural environment; or it can be secondary (that is, synthetic), meaning that it is intellectually constructed from focal visual, verbal, or other symbolic data, such as that presented by flight instruments. While the former type of orientational percept is essentially irrational (not subject to analysis and interpretation) and involves largely preconscious mental processing, the latter type is rational and entirely conscious. A locational percept, to be distinguished from an orientational percept, is a sense of one's position and motion in (as opposed to relative to) the plane of the earth's surface. An accurate locational percept is. achieved by finding one's position on a map or knowing one's latitude and longitude. Spatial disorientation is a state characterized by an erroneous orientational percept, that is, an erroneous sense of one's position and motion relative to the plane of the earth's surface. Geographic disorientation, or "being lost," is a state characterized by an erroneous locational percept. These definitions together encompass all the possible positions and velocities, both translational and rotational, along and about three orthogonal earth-referenced axes. Orientation information includes those parameters that an individual on or near the earth's surface with eyes open can reasonably be expected to process accurately on a sunny day. Lateral tilt, forward-backward tilt, angular position about a vertical axis~ and their corresponding first derivatives with respect to time are the angular positions and motions included; height above ground, forward-backward velocity, sideways velocity, and up-down velocity are the linear position and motions included. Absent from this conceptual collection of orientation information parameters are the location infor- ACCEPTED FOR PUBLICATION 4 August

2 298 mation parameters: linear position dimensions in the horizontal plane. In flight, orientation information is described in terms of flight instrument-based parameters (Figure 1). Angular position is bank, pitch, and heading; and the corresponding angular velocities are roll rate, pitch rate, and turn rate (or yaw rate). The linear position parameter is altitude; and the linear velocity parameters are airspeed, slip/skid rate, and vertical velocity. Inflight navigation information is comprised of linear position dimensions in the horizontal plane, such as latitude and longitude or bearing and distance. Air Force Manual 51-37, Instrument Flying, categorizes flight instruments into three functional groups: control, performance, and navigation. In the control category are the parameters of aircraft attitude (that is, pitch and bank) and engine power or thrust. In the performance category are airspeed, altitude, vertical velocity, heading, turn rate, slip/skid K. K. Gillingham rate, angle of attack, acceleration (G loading), and flight path (velocity vector). The navigation category includes course, bearing, range, latitude/longitude, groundspeed, time, and similar parameters useful for determining location on the earth's surface. This categorization of flight instrument parameters allows us to construct a useful operational definition of spatial disorientation: it is an erroneous sense of the magnitude or direction of any of the aircraft control and performance flight parameters. Geographic disorientation, in contrast, is an erroneous sense of any of the aircraft navigation parameters. Although it might at first appear that inclusion of some of the control and performance parameters (for example, power) as parameters of orientation stretches the definition of SD too far, we would argue that knowledge of each of the aircraft control and performance parameters gives the pilot additional, critical information with which to assess the dynamic "state" of the ANGULAR LINEAR AXIS POSITION VELOCITY POSITION VELOCITY x Bank Roll rate * y Pitch Pitch rate *. Airspeed Slip/skid rate z Heading Turn rate Altitude Vertical velocity * Navigation parameters Figure 1. Flight instrument-based parameters of spatial orientation. Spatial disorientation (SO) Is a state characterized by an erroneous sense of the magnitude or direction of any of these parameters.

3 Spatial Disorientation in USAF aircraft, which assessment is the object and essence of spatial orientation in flight. The utility of these operational definitions of spatial and geographic disorientation is that they can readily be understood by pilots, flight surgeons, and other parties investigating aircraft mishaps. If the answer to the question, "Did the pilot not realize his actual pitch attitude and vertical velocity (and/or other control or performance parameters)?," is "Yes," then it is obvious that the pilot was spatially disoriented, and the contribution of SD to the sequence of events leading to the mishap is clarified. Sometimes aircrew tend to be imprecise when they discuss 'spatial disorientation, preferring to say that they "lost situational awareness" rather than "became disoriented," as though experiencing SD stigmatizes them. Situational awareness involves a correct appreciation of a host of conditions, including the tactical environment, location, weather, weapon capability, own capacities, administrative constraints, and so on, as well as spatial orientation. Thus,' if the situation about which a pilot lacks awareness is the magnitude or direction of any aircraft control or performance parameter, he or she has spatial disorientation, specifically, as well as loss of situational awareness, generally. Sources of SD Although it is beyond the scope of this paper to discuss in detail the hows and whys of SD in flight, I shall provide a conceptual outline of the phenomena that contribute to its development. For further information on the technical aspects of this topic, the reader is referred to USAFSAM Technical Report 85-31, Spatial Orientation in Flight (1). \Ve divide the illusions that result in SD into two types: visual and vestibular. The visual illusions are further divided into those resulting primarily from the ambient mode of visual functioning and those resulting from the focal mode of visual functioning. The major illusions in flight resulting from ambient (that is, involving the entire visual field) visual 299 functioning are the erroneous sensations of pitch and bank attitude caused by false horizons, such as shorelines, and sloping surface planes, such as cloud formations. Visually induced erroneous sensations of linear or angular self-motion, known as vection illusions, also are a result of ambient visual functioning. Absence of ambient visual orientation cues can also contribute to SD: the "black hole" and "whiteout" conditions are notorious for causing pilots to misjudge their flight parameters during approaches at night and over snow-covered terrain, respectively. The illusions resulting from focal (that is, involving the central visual field) visual functioning include such phenomena as misjudging height above terrain because the local structures, vegetation, or other cultural features are not the same size as those with which the pilot is familiar (size constancy), or misjudgingapproach slope'because the local runway is built on, rising terrain and presents an image shape different from that with which tnepilot is familiar (shape constancy). Sometimes absence of focal visual orientation cues can result in illusions: featureless terrain or glassy 'Water can, cause pilots to overestimate,their ;height above the surface of the earth. These,ex: amples of forms of visually induced SD.. are by no means comprehensive, but they are representative. The vestibular illusions are divided into those resulting from effects of semicircular duct function, those resulting from effects of otolith organ function, and those resulting from combined effects. The semicircular ducts within the semicircular canals are responsible for detection of angula!' motions of the head, Vestibular illusions of semicircular-duct origin are known as somatogyral illusions, which are false sensations of rotation (or absence of rotation) that result from misperceiving the magnitude or direction of actual rotation. The "graveyard spin" and the vestibular coriolis effect are examples of somatogyral illusions. The otolith organs (utricle and saccule) within the vestibule detect linear motions of the head and the direction of the force of gravity. The vestibular illusions of otolith-organ origin are the somatogravic illusions and the G-excess

4 300 effect. A somatogravic illusion is a false sensation of body tilt that results from misperceiving as vertical the direction of a nonvertical gravitoinertial (G) force. Examples of this type of illusion are the sensation of nose-high pitch that pilots often experience at night during forward acceleration of their aircraft, and the sensation of being inverted that sometimes results when an aircraft is leveled off abruptly after a climb in weather. The G-excess effect, an exaggerated sensation of body tilt that occurs when the sustained G load is greater than the normal 1 G, is believed to cause pilots to perceive falsely that they are underbanked during level steep turns and to overbank and inadvertently descend as a result. The G-excess effect is also responsible for an illusion of pitching up after the arrest of a sustained descent, and can result in the pilot's re-establishing an unperceived descent. The most common form of SD in flight, the "leans," and its often fatal consequence, the "graveyard spiral," both result from a combination of somatogyral and somatogravic illusions. A general principle that must be emphasized regarding vestibular illusions is that they are suppressed by ambient vision. In other words, vestibular illusions can become manifest when visual reference to the natural environment is compromised by in-cockpit duties, G-induced tunnel vision, and the like, or when the natural environment itself is relatively devoid of orienting cues, as during night and weather flying. Sensory systems other than the visual and vestibular. can contribute to SD. The cutaneous mechanoreceptors that provide sensations of touch and pressure normally help orient us to the direction of the force of gravity and can signal transient linear accelerations. But in flight, this sensory system is subject to making the same errors that the vestibular system makes, and tends to reinforce rather than dispel vestibular illusions. The proprioceptive end-organs in the muscles, tendons, and joints can also provide neural information that contributes to SD in flight. Even auditory cues can be used on the earth's surface to provide some degree of spatial orientation in the absence of vision, but in flight the virtual absence of such cues may contribute to SD. K. K. Gillingham So far I have discussed how a pilot's sensory systems are inadequate to assure accurate orientation at all times and under all conditions of flight. Both the reflex and voluntary motor systems can also contribute to the development and persistence of SD. Postural reflexes of visual or vestibular origin can make a pilot lean his body in the direction of a falsely perceived upright position (hence the ~erm "leans"). Occasionally, prolonged vestibula-ocular reflex activity, that nystagmus, is elicited by sustained or repetitive. highamplitude, angular acceleration in flight. These involuntary, rapid oscillations of the eyes can prevent the pilot from obtaining a clear view of either the flight instruments or the natural environment, thus making recovery from SD very difficult during, for example, uncontrolled spinning or rolling of the aircraft. More important from a practical standpoint than reflex motor activity is disorientation-induced bias of voluntary motor activity. The "giant hand" phenomenon, in which a disoriented pilot feels his correcting flight control inputs to be actively countered, as if by a giant hand, probably results from such a m.otor bias. The influence of preconscious, erroneous, vestibular orientation information on the pilot's conscious, but automatically executed, motor acts to maintain equilibrium apparently can elicit in the pilot the sensation of an aircraft flight control systern malfunction. Types of SD We distinguish three types of spatial disorientation in flight: Type I (unrecognized), Type II (recognized), and Type III (incapacitating/ uncontrollable). In Type I, the pilot does not consciously perceive any of the manifestations of SD; that is, he experiences no disparity between natural and synthetic orientational percepts, has no suspicion that a 11ight Lrlstr...!ment (for example, attitude indicator) has malfunctioned, and does not feel that" the aircraft is responding incorrectly to his control inputs. To distinguish Type I SD from the others, and to emphasize its insidiousness, some pilots and

5 Bpatial Disorientation in USAF aerospace physiologists call Type I SD "misorientation. " I believe use of this term creates more confusion than clarification. In Type II SD the pilot consciously perceives some manifestation of disorientation. He may experience a conflict between what he feels the aircraft is doing and what he determines from the flight instruments that it is doing. Or he may not experience a genuine conflict, but merely conclude that the flight instruments are lying to him. He also may feel that the aircraft is attempting to assume a pitch or bank attitude other than the one that he is trying to establish-that is, he may experience the giant hand. Type II SD is the kind to which pilots are referring when they use the term "vertigo," as in "I had a bad case of vertigo on final approach." Although Type II SD is ;Iabeled "recognized," the pilot may not actually realize he has SD: he may only realize he has a problem ascertaining or controlling the performance of his aircraft, -without the shghtest idea as to why. With Type III SD the pilot experiences an overwhelming - that is, incapacitating - physiologic response to physical or emotional stimuli associated with SD. He may have nystagmus to such a degree that reading the flight instruments -is impossible. Or he may have such a strong expression of the giant hand that control of the aircraft is impossible. He may even be so incapacitated by fear that he is unable to make a rational decision - he may freeze on the controls. The important feature of Type III SD is that the pilot is disoriented and most likely knows it, but can't do anything about it. Fortunately, this type of SD is relatively uncommon. The Cost of SD Spatial disorientation is the largest single cause of Class A (major) aircraft mishaps attributable to operator error in the United States Air Force (USAF). Data from the Air Force Inspection and Safety Center (Freeman JE, personal communication, 1990) indicate that, over the 10-year period from 1980 through 1989, approximately one-eighth of all 301 USAF Class A mishaps were reported to have involved SD as a causal or contributory factor,accounting for an average of 8 such mishaps per year. If only the operator-error mishaps involving fighter lattacklreconnaissance aircraft are considered, SD was implicated in one-fourth of these, accounting for 5 or 6 per year. If we look at only the operator-error mishaps involving the USAF's frontline fighters, the F-15 and F-16, one-third of such mishaps - 2 or 3 per year - were at least partially due to SD. The annual dollar, cost of SD to the USAF is currently on the order of $100 million; but occasionallossesolparticuiarly expensive aircraft have resulted in much higher figures.in some years. Even more unfortunate is the fact that,onaver.age, 12peQ~ pie are killed each year in SD"\related USAF aircraft mishaps~.although I don't have current statistics on mishaps involving US Army and US Navy aircraft, one'can safely assume that their losses.resulting from SDar.e;-similar to those experienced by the USAF. When one considers the number of air forces in the world, and the magnitude of commercial and general aviation activity, it is easy to arrive at an estimate that SD wastes billions of dollars worth of resources every year. One problem with the USAF mishap statistics presented above is that they are conservative, representing only those mishaps in which SD was stated to be a possible or probable factor by the safety investigation board responsible for determining and reporting the facts relating to each mishap. In actuality, many mishaps resulting from SD were not identified as such because other factors - such as distraction, task saturation, and poor crew coordination - initiated the chain of events resulting in the mishap; these factors were considered more relevant or more amenable to correction than the SD that followed and ultimately caused the pilot to fly the aircraft into the ground or water. The statistics from the Air Force Inspection and Safety Center on USAF Class A aircraft mishaps from 1980 through 1989 (Table 1) reveal that "loss of situational awareness" (that is, channelized attention, distraction, task saturation, and the like) was responsible for 'over of all such

6 302 K. K. Gillingham Table USAF Class A Aircraft Mishaps-as Categorized by Safety Investigation Boards Operations SO LSA SO/LSA Total related related related related Mishaps Fatalities Cost (US $) 4,452M 2,558M 539M 2,012M 2,045M so = spatial disorientation. LSA = loss of situational awareness (channelized attention, distraction, task saturation, and so on). mishaps, and for three-fourths of the Class A mishaps resulting from operator error. It is apparent that the great majority of these "loss of situational awareness" mishaps would not have happened if the pilots had continuously and correctly assessed their pitch/bank attitude, vertical velocity, and altitude-that is, if they had not been spatially disoriented. Thus, we can infer that SD causes many more USAF aircraft mishaps than the SD-specific incidence statistics would lead us to believeprobably two to three times as many. Regarding the fractions of the SD mishaps for which the various types of SD are responsible, the conventional wisdom is that more than halfof SD'mishaps involve Type I, most of the remainder involve Type II, and very few involve Type III. The same wisdom suggests that the source of the SD is visual illusions in about half of the mishaps, and vestibular / somatosensory illusions in the other half, with combined visual and vestibular illusions accounting for at least some of the mishaps. An analysis of USAF aircraft mishaps in 1988 in which SD was suspected by the investigating flight surgeon revealed that an 8 involved Type I SD; 2 apparently resulted from visual illusions, 3 from vestibular illusions, and 3 from mixed visual and vestibular illusions (2). Major Factors The numerous factors that can contribute to the development of SD in flight, including environmental, mission, aircraft, and pilot factors, are discussed in USAFSAM-TR (1). I shall emphasize here some'factors that seem to have become especially'relevant recently. These factors fall into two areas of concern: task loading and training. In the task loading area, modern aircraft designs and mission requirements have substantial impact. First, the phenomenal ability of current high-performance aircraft to roll, pitch, accelerate, gain or lose altitude, and otherwise change spatial orientation parameters very quickly, presents a significant challenge to the pilot to maintain a continually accurate assessment of those parameters. The improved visibility afforded the pilot by the great expanse of clear canopy in modern fighter aircraft has one negative feature: it makes the pilot more susceptible to orientational illusions of ambient visual origin. Many modem military aircraft have electronic systems that enhance the pilot's ability to fly at night and in instrument weather conditions.. This enhanced capability, of course, allows pilots to spend more time in environmental conditions that are conducive to the development of SD; Perhaps the most important and controversial aircraft design consideration relating to SD is location and type of flight instruments. Whereas the previous generation of fighter / attack aircraft had a relatively large attitude indicator and other primary flight instruments located directly in front of the pilot on the instrument panel, the recent trend has been to use such valuable instrument panel space for other than the primary flight instruments; that is, it is used for weapons delivery or navigation functions, and the instruments for spatial orientation are relegated to smaller spaces and more remote locations on the panel. The result is that the mental work expended by the pilot to maintain spatial orientation is in-

7 Spatial Disorientation in USAF creased. Furthermore, the head-up display (HUD) has evolved from purely a targeting aid to an instrument that can be used for flight control, performance, navigation, and weapons delivery functions. But because of certain deficiencies in current HUD symbologies, spatial orientation is generally more difficult to acquire and maintain with the HUD than it is with the conventional panel instruments. As many pilots have elected to use the HUD as their primary flight reference (indeed, some aircraft require using the HUD as such), the potential for SD has become greater rather than less in the wake of this new display technology., Another very important factor that has increased task loading of the pilot, and thereby increased the potential for SD, is the "greater use of single-crew as opposed to dual-crew fighter lattack aircraft. Although he has considerable electronic help, the modern pilot is responsible for at least as much information as two aircrew used to be, so he has less time available to dedicate to spatial orientation. Partly as a result of improved avionics capability and partly as a result of changes in the air combat environment, mission requirements are now more conducive to SD mishaps than was formerly the case. There is substantially more emphasis now on an ability to operate at night, under the weather, and in weather. There is also increased emphasis on low-level, high-speed operations, wllich certainly raise the likelihood of disaster whenever SD does occur, because of the decreased margin for error in control of flight parameters. We also believe that less than optimal aircrew training is an important factor in the current SD problem. Although physiological training has addressed the topic of SD in lec., tures to aircrew for many years, more could and should be done to ensure that pilots are keenly aware of the hazard that SD represents. We feel it would be especially productive to give every pilot a realistic, personal demonstration of some of the common manifest a- tions of SD in both a physiological training and a flying training setting. There is also a need to ensure that all pilots have the instrument flying skills necessary to fly safely in all 303 environmental conditions to which they are likely to be exposed. Pilots receiving all their flight training in the sunny American Southwest, and getting almost all their instrument flying experience in instrument procedures trainers, flight simulators,andaircraft without vision-restricting devices, are not always well prepared for operational flying in the gloomy murk of the northern climes to which they may be deployed. It is also apparent that at least some pilots need additional instruction and/or practice in task management and prioritization, especially during low-level flying. SD Countermeasures' Research and Developm~nt'.,.+,.... Because t.heforegoingfactqf;'s conspire,to maintain or even increase, the, frequency of occurrence of SD mishaps in' the" USAF" we believe a significant R&D investment to try to reduce SD..;related attrition 'ofaircraft and aircrew is clearly appropriate. An eventual solution to the SDproblem, other than remotely piloted or' even autonomous vehicles, is the use of acomputer-generated, wide-field-of-view, helmet-mounted or cockpit-based visual display.of the essential orienting features of the real world, which the pilot can use to maintain his spatial orientation in the natural manner. But realization and implementation of such a display is a distant goal, and interim measures to prevent SD mishaps must be sought. For the near term, the only practicable approach is to improve SD awareness training for pilots; for the moderately far term., flight instruments providing more efficacious displays of spatial orientation information can be fielded. To those ends, we have formulated a three-pronged R&D attack on the SD problem: 1) elucidate the basic psychophysiologic mechanisms of spatial orientation and disorientation, 2) developsd awareness trairjng methods for both ground-based and inflight application, and 3) design and test concepts for flight instruments and displays that contribute to enhanced spatial orientation.

8 304 Mechanisms Research The research to elucidate the mechanisms of spatial orientation and SD is divided into four thrusts - three relating to the sensory systems that are or can be used to generate orientation information, and one relating to determining the operational factors that contribute to SD mishaps. Research on visual orientational mechanisms is extremely important because so many mishaps are caused visual illusions, and because several types of solution to the SD problem depend onvisual orientation information. An understanding of the principles of visual spatial orientation and visually induced SD is essential for the development of demonstrations of visually generated forms of SD in ground-based training devices and curricula, and for the design of more effectively orienting symbology in HUDs and helmet-mounted displays' (HMDs). Research on vestibular orientational mechanisms has been carried out for decades, and much is known about vestibular causes of SD. But,,' a number of recent SD-related aircraft mishaps suggest that more needs to be known about the perceptual responses generated by the otolith organs, especially tinder conditions of sustained, higher-than-normal G forces. Many of these recent mishaps appear to have been caused by the vestibular G-excess effect, and quantification of this phenomenon is necessary (3). Data generated on the G-excess effect and other vestibular mechanisms of SD will be used primarily in the development of ground-based and inflight training methods, the object of which will be to improve pilots' awareness of how and when SD can occur and how they can either prevent it or cope with it. Auditory orientation research has application in the development of acoustic displays of orientation information. The new technology of synthetic auditory localization can possibly be harnessed to provide the pilot with an alternate sensory channel for accurate spatial orientation information when the visual channel is saturated with information needed to satisfy nonorientational demands. Consultative support to safety investigation boards in aircraft mishap investigations pro- K. K. Gillingham vides researchers with valuable insight into the operational factors that contribute to SD mishaps, as do retrospective examinations of mishap data that produce statistical evidence of problem areas and trends. Such information is critical, not only for the development of training methods and materials, but also for providing direction in conceptualization and design of flight instruments and displays to help prevent SD in real situations. Training R&D Both ground-based and inflight SD awareness training methods need to be further developed and refined. The ground-based methods center on the Advanced Spatial Disorientation Demonstrator (ASDD), which we hope will replace the Vista Vertigon as a tool for USAF aerospace physiologists to use in demonstrating representative visual and vestibular orientational illusions to pilots and in training pilots to recognize and cope with SD. The current plan is to acquire an ASDD, develop prototype training curricula for use with it, thoroughly examine its potential for enhancing SD training, and then make recommendations regarding whether and how best to implement ground-based SD training based on ASDD technology. Concurrently, we need to develop methods that instructor pilots and instrument check pilots can use to demonstrate relevant vestibular forms of SD consistently and reliably in flight. Both the ground-based and in flight training methods developed will require validation to ensure that the training as implemented is likely to be effective. ASDDbased training, in particular, will have to be examined not only for efficacy and for possibly counterproductive approaches, but also for untapped training potential in areas not initially addressed (for example, task management/prioritization training). Instruments/Displays R&D The three types of display to which we confine our attention are HUDs, HMDs, and an acoustic display. Although tactile displays are

9 Spatial Disorientation in USAF a possibility, the risk/gain ratio for such technology appears rather high, and we feel our resources would be better invested in more thorough development of the relatively conservative visual and acoustic media than in examination of a tactile display. Improvements in HUDsymbology will continue to be suggested and tested, both in the laboratory and in flight, with the objective of enabling the HUD to serve as a singlesource primary flight reference system, thereby assuming that function from the conventional instrument panel indicators. Physiologic principles relating to the three-dimensional structure of visual attention and to eye movement control, elucidated in the studies of visual orientational mechanisms, are applied in the development of improved HUD symbology. Candidate elements of HUD symbol sets are rigorously compared for relative efficacy in laboratory simulations of flight, and those exhibiting superior ability to help pilots mafntain their orientation and control over flight parameters are incorporated into a draft USAF standard HUD symbol set (4). Inflight validation of such standardized HUD symbology is then carried out. Our efforts to improve and standardize HUD symbology primarily support the USAF Instrument Flight Center in its mission to promote safe instrument flying. HMD technology will eventually reach fruition, and at some point will probably eliminate SD. R&D efforts to advance such technology should be strongly supported, not only f0r the primary purpose of making targeting and threat warning easier, but also to help the pilot stay spatially oriented at all times. AI,. though HMD R&D remains primarily in the domain of weapons delivery avionics; the tremendous potential for HMDs to provide readily assimilable spatial orientation information obligates us to ensure that HMDs develop also as vehicles for primary flight information. Computer image generation technology already allows the creation of a synthetic total spatial environment, but the actual display of the synthesized environment at optical infinity on or inside the visor of a suitable helmet remains a technical challenge. While HMD hardware development progresses, we will 305 provide appropriate design guidance and will evaluate candidate spatial orientation symbologies for HMDs. An interesting alternative approach is the use of sound to provide accurate orientation information. We are currently evaluating the potential of a stereophonic acoustic orientation instrument (AOI) to help pilots maintain aircraft control when their vision is either otherwise occupied or temporarily incapacitated. The AOI has been demonstrated to be effective, both in a flight simulator and in a research aircraft, in allowing blindfolded pilots to maintain control of aircraft bank angle and, to a lesser extent, vertical velocity (5). The next stage of AO I development will incorporate synthetic three-dimensional sound, that is, auditory localization (as opposed to the auditory lateralization presently employed). With this 3-D audio technology, an acoustic attitude indicator featuring a sonic sky-pointer is theoretically possible. Although the AO I, is presently a stand-alone instrument that could be developed and transitioned as such, It is best considered 'an intertnediatestep representing technology that will be integrated into the HMDs of the future. The Distant Future Earlier I raised in passing the idea that remotely piloted or autonomous vehicles could eliminate the SD problem. Such an idea seems much less incredible today than it did even a few years ago. With the imminent realization of high-technology navigation aids - namely) the satellite-based Global Positioning System and the Defense r-v1apping digital terrain data - sufficient real-tiltle information will be available for aircraft with advanced sensors and onboard computers to function autonomously from takeoff to landing. This capability will, I believe, revolutionize aerial operations: the pilot's main function will not be to fly the airplane but will be to monitor information systems and to assume control in case of a system failure. Inevitable improvements in systems reliability will eventually eliminate the need for pilots altogether; and the SD problem, which has its origin in the pi-

10 306 lot's human frailties, will thus disappear. We will be well into the twenty-first century, however, before this final solution to the SD problem materializes; and we cannot passively wait for it. Conclusion Our approach to solving the SD problem is similar to that recently employed tackthe problem of G-induced loss consciousness (G-LOC) in f1ight: we believe that research on underlying mechanisms is productive, that aircrew training will be useful in the short run, and that hardware improvements will eventually provide substantial additional protection. But the SD problem is likely to be a more difficult problem to solve than the G-LOC problem. First, SD involves the complexity of multiple, interactive, perceptual processes, while G-LOC is basically a problem of cardiovascular hydraulics. Second, the environment in which SD is likely to occur is less well defined than is the case with G-LOC. The amount of time pilots fly in conditions in K. K. Gillingham which SD can occur is an appreciable percentage of their flying time, and the fact that a particular condition can precipitate SD isn't always obvious to the pilot. During their relatively brief exposures to high G loads, on the other hand, pilots should be well aware of the imminent danger of G-LOC and should always be prepared to counter it. Third, SD has been recognized as an aviation safety problem for over half a century, and we have become inured to the relatively high percentage of mishaps caused by SD. But G-LOC, although it is responsible for only a small fraction of the number of USAF aircraft mishaps due to SD (and almost no nonmilitary aircraft mishaps), is a novel, sensational cause of mishaps; accordingly, a great deal of attention and commensurate resources have been dedicated to solving the G-LOC problem. Nevertheless, progress is being made toward mounting an attack on SD. There has been a fivefold increase in our investment in SD countermeasures R&D over the past five years. As a result, we have gained the necessary momentum to have an impact on the problem in the not-toodistant future. REFERENCES 1. Gillingham KK, Wolfe JW. Spatial orientation in flight. USAFSAM-TR Brooks AFB, Texas: USAF School of Aerospace Medicine; 1986 December. 2. Lyons TJ, Freeman JE.. Spatial disorientation (SD) mishaps in the US Air Force-1988 [abstract], Aviat Space Environ Med. 1990;61: Lyons TJ, Gillingham KK, Thomae C, Albery WB. Low-level turning and looking mishaps. Flying Safety ct: Weinstein LF, Ercoline WR. The standardization of head-up display symbology for the reduction of spatial disorientation [abstract]. Aviat Space Environ Med. 1992;63: Gillingham KK, Teas DC. Flight evaluation of an acoustic orientation instrument (AOI) [abstract]. Aviat Space Environ Med. 1992;63:441.