Click for next page ( 146


The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement



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 145
VISION PROBLEMS IN LOW-ALTITUDE, HIGH-SPEED FLIGHT Introductory Remarks by the Chairman James W. Miller Engineering Psychology Branch Office of Naval Research The papers presented in this Section describe current work re- lated to the visual aspects of low-level, high-speed flight, and include such topics as geographic orientation, visual acquisition and tracking, tactical strike missions, and low-level problems associated with helicopters. The concept of flying at extremely low altitudes in order to penetrate enemy territory is by no means a new one. However, recent advances in radar defense systems and the accuracy of guided-missile systems have made it impera- tive that attacking aircraft avoid early detection, if at all possible, in order to insure survival. There are several types of missions which require flying at these extremely low altitudes. Such mis- sions are usually concerned with visual reconnaissance, ground support, special-weapons delivery, or fire control. The Low-Altitude High-Speed (LAHS) mission requires pre- cise location and identification of ground points with precise heading, air-speed, and altitude. Such navigation is a combina- tion of dead reckoning, and visual and radar navigation using ground features as an aid in the identification of targets and checkpoints. Terrain-avoidance radar and other airborne dis- play systems are being developed in order to aid the pilot in such missions. Recent evidence indicates, however, that, in large numbers of flights, the pilot becomes so geographically disoriented that the goal is not achieved at all and the entire mission must be aborted. It is hoped that these papers will clarify those problems which currently face pilots flying such missions. 145

OCR for page 145
GEOGRAPHIC ORIENTATION DURING LOW-ALTITUDE FLIGHT1 James J. McGrath Human Factors Research, Incorporated Los Angeles, California In low-altitude flight, the pilot is burdened with a large number of information and performance requirements. One of these requirements, that he maintain an awareness of his navigational position, has become known as "geographic orientation." (Geo- graphic orientation should not be confused with the more familiar spatial orientation, which generally refers to the pilot's aware- ness of the attitude of his aircraft.) Before undertaking any extensive research on geographic orientation, it was first necessary to determine the need for re- search in this area and the directions it should take. Therefore, in June, 1963, a "Phase 1" effort began. It was essentially a problem analysis effort, rather than a problem solving effort. The purposes of the research were: first, to determine whether pilots under operational conditions become lost frequently enough to constitute a significant cause of mission failure; secondly, to determine the role for research and to establish the priority of research problems within this area: and, finally, to evaluate pos- sible experimental techniques to be used in solving the research problems. The inquiry was limited during this phase to those missions in which the pilot must fly at low altitudes under visual flight rules. 1. This research is being supported under Contract Nonr 4218(00) by the Joint Army-Navy Aircraft Instrument Research project, a coopera- tive effort of the Office of Naval Research, The Bureau of Naval Weapons, and the U.S. Army Material Command. 146

OCR for page 145
THE OPERATIONAL PROBLEM The first task of the project was to determine the magnitude of the problem of geographic disorientation in present-day aviation operations. The guiding principles in carrying out this task were: (a) the magnitude of the problem of geographic disorientation should be documented by objective data, rather than by personal opinion: (b) the data should come from several different sources, rather than from a single source: and (c) the data should be cur- rent to reflect conditions found in present-day aviation. Aircraft Accidents An obvious indication of the magnitude of the problem would be the number of aircraft accidents attributable to geographic dis- orientation. Therefore, aircraft accident records were obtained from the Air Force, Navy, Army, and Civil Aeronautics Board and were examined to determine the role of geographic disorien- tation in aviation safety. Fig. 1 shows the number of aircraft completely destroyed and the number of lives lost in major aircraft accidents attributable to geographic disorientation in military aviation during the years 1958 to 1962. The tabulation does not include accidents in which the aircraft sustained reparable damage, nor those in which the crew were only injured. Accidents in which geographic disorien- tation played only a minor or inconsequential role were also excluded. In civilian aviation during the three-year period 1959-1961, a total of 243 accidents resulted from the pilot becoming lost under Visual Flying Rules (VFR) conditions. In these accidents 41 per- Q60 u 550 K « 40 u !» u. 2 20 u 5 10 ^ USAF USN USA 60 50 2 40 20 10 0 USAF USN USA FIG. 1. Number of aircraft destroyed and men fatally injured in aircraft accidents attributable to geographic disorientation in military aviation, 1958-1902.

OCR for page 145
sons were killed. Many pilots became lost because they continued VFR into Instrument Flying Rules (IFR) weather conditions. Dur- ing the same three-year period, 613 accidents resulted from this type of geographic disorientation, and 365 of these involved fa- talities. Geographic disorientation was a contributing cause of 6.7 per cent of all general aviation accidents during the period studied. Although aircraft accidents resulting from geographic dis- orientation have taken a heavy toll of aircraft and human lives, such aviation safety data do not accurately reflect the frequency with which pilots have become lost. The vast majority of pilots who become lost survive the experience, and accident statistics describe only a minor aspect of the problem. Personal Experiences of Military Aviators Another source of data was the pilot himself. A total of 72 Navy, Marine Corps, and Army pilots was asked to describe their most serious personal experiences with geographic disorientation. All but four pilots described at least one instance of becoming lost in flight; the majority said they had had many such experiences. Forty-five of these descriptions of critical incidents were re- corded in detail. One of the items of information recorded was the date on which the incident occurred. Fig. 2 shows the distribution of critical incidents by calendar year, and Fig. 3 shows the distribution by month within the year 1963. Note that the most frequent category was 1963; and within that year, most of the incidents occurred during July—the month 63 62 61 60 59 58 57 56 55 54 53 FIG. 2. Forty-five critical in- cidents of geographic disori- entation reported by military aviators, distributed by calen- dar vcar. 148

OCR for page 145
JULY JUNE MAY APR MAR FEB MONTH. 1963 FIG. 3. Twenty-two critical incidents of geographic disorientation re- ported by military aviators, distributed by calendar month during 1963. during which the interviews were conducted. One may draw either of two conclusions from the data: that an epidemic of geo- graphic disorientation occurred in July, 1963, or that geographic disorientation is a relatively common occurrence among military aviators. When asked to relate his most serious personal experi- ence of being lost, the pilot simply relates his most recent one— usually an experience which has occurred within the past few months. Flight Assists to Lost Civilian Aviators The large number of civilian aviators who become lost is indi- cated by the Flight Assist Reports issued by the Federal Aviation Agency (FAA). In 1962, a total of 1,492 assists was given by ground control operators to pilots flying the federal airways system. Of this number, 1,270 assists were given because the pilots were lost. In other words, of all the pilots who required assistance for all reasons, 88 per cent required assistance be- cause they were geographically disoriented. This is a conserva- tive number, according to FAA officials. They believe that assist- ing lost pilots to reorient themselves is such a commonplace event in air traffic control that a great many such incidents go unreported. Geographic Disorientation on Low-Altitude Missions To get closer to the heart of the military problem, attention was turned to the low-altitude attack training missions which are now being flown over the Sandblower courses. (Sandblower courses are those routes designated for low-level training and extend from the Northwest to the deserts.) These missions are assumed to be representative of operational combat missions flown by light attack aircraft. Mission critiques of almost one thousand Sand- 149

OCR for page 145
blower flights were analyzed. Based on the statements of the chase pilots, who evaluated the missions, Fig. 4 shows the inci- dence of geographic disorientation on these flights. Ten per cent of the missions failed completely because the pilot became lost. In another 17 per cent of the missions the pilot became lost, but eventually reoriented himself and found the target area. In these cases the mission was compromised in many ways, and under combat conditions might have failed. The remaining 73 per cent of the missions were categorized as "O.K." Many of these mis- sions failed too, but not because of geographic disorientation. FIG. 4. Summary of data obtained from 959 flight critiques of low-level navigation training missions. It is important to point out that the "abort" and "recover" percentages are minimum figures, in that many cases of disori- entation went unreported, and any case that was doubtful, or that had insufficient information, was thrown into the "O.K." category. Records of Individual Pilots To clarify the data obtained from the flight critiques, a study of individual differences in low-altitude navigational performance was made. A "probability of disorientation" was computed for each pilot, based on the proportion of his flights on which he be- came geographically disoriented. The frequency distribution shown in Fig. 5 was plotted from the resulting data. The distri- bution includes 126 pilots, each of whom flew between six and 150

OCR for page 145
PROBABILITY OF DISORIENTATION FIG. 5. Frequency distribution of disorientation incidents showing in- dividual differences in geographic orientation performance. eleven low-altitude navigational missions. Individual differences ranged from becoming lost on no mission to becoming lost on every mission. For the reasons mentioned earlier, these data are conservative estimates of the frequency of geographic dis- orientation on these missions. The data can be interpreted more meaningfully in the form shown in Fig. 6, which is a cumulative distribution of the data shown in Fig. 5. The abscissa represents probability of dis- orientation based on the percentage of missions on which the pilot became lost. The ordinate represents the percentage of pilots who exceeded that probability. By drawing a line horizon- tally from the 50 per cent point on the ordinate to the cumulative distribution function, and, thence, vertically to the abscissa, for example, it can be seen that one-half of the pilots became geo- graphically disoriented on at least 23 per cent of their missions. By intersecting the function at another point, as another example, it can be seen that 15 per cent of the pilots became disoriented on about one-half of their missions. When individual differences in performance are of the magni- 0 .1 .2 .3 .4 .5 .6 .7 .8 .9 1.0 PROBABILITY OF OISORIENTATION (Pd> FIG. 6. Percentage of pilots exceeding a given probability of geographic disorientntion (N = 126) 151

OCR for page 145
tude shown here, it is always valuable to know the reliability of such differences. To estimate the reliability of individual dif- ferences in geographic orientation performance the correlation was computed between the instances of disorientation that oc- curred on each pilot's even-numbered flights with those which occurred on his odd-numbered flights. The correlation coeffi- cient was 0.56. This is a remarkably high reliability coefficient in view of the nature of the data from which it was derived. Conclusions The results of the initial investigation showed that aircraft acci- dents caused by geographic disorientation take a significant toll of aircraft and human lives. But the reports of critical incidents by military pilots and the large number of flight assists to lost civilian pilots indicate that geographic disorientation is far more prevalent than indicated by accident data. If the training opera- tions that are conducted on the Sandblower courses by light attack aircraft are at all representative of low-altitude combat missions, it can be expected that a substantial number of such missions will fail, or will be seriously compromised by geographic disorienta- tion. Further, the problem is not confined to a small number of disorientation-prone pilots, but appears to be a general problem encountered by the majority of pilots. One is compelled to con- clude that the answer to the original question was "yes": geo- graphic disorientation does occur frequently enough in present- day aviation operations to affect significantly the success of those operations. THE ROLE FOR RESEARCH It does not suffice to point out the existence of a problem and to suggest vaguely that something be done about it; therefore, the next task of the project was to delineate the role for human per- formance research in the study of geographic orientation. Statistical Description of the Problem Situation The first step was an attempt to provide a statistical description of actual incidents of geographic disorientation. The data were derived from accounts of critical incidents of geographic dis- orientation as related by 72 military pilots. Fig. 7 shows some of the functions that were obtained. Fig. 7(A) shows the frequency of disorientation as a function of the number of miles from the point of the take-off to the point at which the pilot realized he was 152

OCR for page 145
(A) 100 200 3OO 400 MILES FLOWN SINCE TAKE-OFF IBI 20 30 40 MINUTES SINCE LAST CHECKPOINT (90) (0 I0 20 30 40 50 DURATION OF DISORIENTATION [MINUTES) (120) (D) (75) mm^^mmm^ 20 30 40 MILES OFF COURSE FIG. 7. Frequency (f) of geographic disorientation in- cidents as function of (A) miles flown since take-off; (B) minutes since last identified checkpoint; (C) dura- tion of disorientation episode; and (D) miles off course during disorientation episode. lost. Disorientation incidents occurred about equally often at all distances from less than 50 miles to more than six hundred miles from the point of origin. In other words, the occurrence of geo- graphic disorientation was not systematically related to distance flown, and it appears that geographic disorientation can overtake the pilot at almost any stage of a flight. Fig. 7(B) shows the frequency of disorientation as a function of the amount of time that had elapsed since the last identified checkpoint. Again, the range was very large. It is particularly interesting to note that a number of pilots became disoriented only a few minutes after they had fixed their positions with an identified checkpoint. Fig. 7(C) shows that the durations of the disorientation ex- periences ranged from about 4 minutes (min) to more than an hour. The median duration was 12 min which, in an aircraft, is a dangerously long period of confusion. 153

OCR for page 145
Fig. 7(D) shows that during these episodes of geographic dis- orientation the pilots were anywhere from 4 or 5 miles to 75 miles from their intended track. The median distance the pilot got off course during these incidents was 20 miles, which is quite a large enough error to compromise most aircraft missions, although at least one incident is known in which the pilot was seven hundred miles off course. The study also disclosed that in one-third of these cases the pilot experienced marked disbelief of some informational source during the disoriented state. That is, he became incredulous of his chart, his instruments, or his preflight planning. Most of the pilots experienced great difficulty in recognizing the fact of dis- orientation, and, having recognized it, experienced marked emotional stress. It was also noted that the pilots used different methods of re- orienting themselves. Of the pilots studied, 27 per cent first searched the chart for possible checkpoints and then attempted to locate these points in the visible field; another 43 per cent first selected terrain features that they could see and then at- tempted to locate these on the chart, while others (30 per cent) used a combination of these methods. In all cases the pilot attempted to relate some feature or combination of features in the visible terrain to features portrayed on his chart. The point of interest here is that the operational procedure recommended by Naval training doctrine is the chart-to-terrain method, the method used by the minority of pilots. A statistical analysis was also made of the aircraft accident data. Although the circumstances surrounding the accidents were diverse, the accidents themselves were easily classified by type. Fig. 8 shows the different types of aircraft accidents that re- sulted from geographic disorientation in military aviation. More than half of these accidents occurred when the pilot believed him- self to be in one position when, in fact, he was many miles away from that position. Unaware of this disorientation, the pilot sub- sequently collided with an elevated portion of the terrain. In the remainder of the cases, the pilot was well aware of the fact that he was lost, but was unable to reorient himself before fuel star- vation occurred. Depending on the circumstances, he either abandoned the aircraft or attempted an emergency landing. In addition to the predominant behavior pattern of failing to recog- nize the fact of disorientation, there was another frequently ob- served pattern. In these cases, the pilots recognized they were lost, but were reluctant to admit this to anyone, and exhausted 154

OCR for page 145
ABANOONED AIRCRAFT 31% FIG. 8. their fuel in an attempt to get out of what they considered to be an embarrassing situation. In civilian aviation these patterns of pilot behavior also occur. But, by far the dominant pattern in civilian accidents caused by geographic disorientation is the inadequacy or complete lack of preflight planning. The General Problem Areas These statistical analyses provided useful background informa- tion on the problem of geographic disorientation, but the identifi- cation of research problems came from an exhaustive study of all available informational sources. Some of the sources have already been described: aircraft accident reports, critical inci- dent reports, flight assist reports, and the critiques of low- altitude missions on the Sandblower courses. Those four sources of information provided data on actual cases of geographic dis- orientation. Two other sources of information were consulted, both dealing with pilot opinion. The two additional studies involved personal interviews conducted with 109 pilots who are presently flying low-altitude missions, and a questionnaire survey of 305 U.S. Army aviators. The large number of specific research re- quirements that were identified from the analysis of all available data has been reported in detail in a technical report (McGrath and Borden, 1963). The following six general problem areas were identified in 155

OCR for page 145
l i i 1 Gfond X g 8 X X X X X X X X X X X X X X X 1 2 g 1 1 1 Six 1 X SIS X X X X X X Ntt X X 1 X X X X X X X X X X X X X X X X X X X X X XX 34 32 30 75 26 24 22 20 IB 6 u U 10 B 6 4 2 * Acquililjon Diltonce (fix l000) Lin* FIG. 15. Frequency of occurrence of acquisition dis- tances for each run. apparent learning. Although it was known that the variance due to run number could affect the rest of the data, it would have been unreasonable to have attempted to make any compensations in the analysis before further test results were obtained. The fact in question is not so much that there was learning, but rather the degree of learning and the point at which the learning stabi- lizes. It is felt that after the second set of target passes by a particular pilot, his learning reaches a plateau. After additional testing has been accomplished, an attempt will be made to sub- stantiate this theory. Although the results of this test were obtained from a small sample size, there was strong support for the fact that target acquisition was affected significantly by the speed of the aircraft. In fact, the distance from the aircraft to the target at acquisition was decidedly less for air speeds in the supersonic region. It is felt that this distance was less to the extent that it put the pilot beyond the critical point in reaction time at which he could make the required tactical corrections for delivery of ordnance on the target. In light of these factors, another test is presently being con- ducted at the Air Proving Ground Center to evaluate training techniques designed to improve the perceptual efficiency of pilots and thus improve their capability for target acquisition and recognition. These techniques involve use of linearly programmed photo- graphs (ranging from low stimulus ambiguity to high stimulus ambiguity) of typical tactical targets which will be presented on a tachistiscopic teaching machine, allowing stimulus presenta- tion speeds of 1 sec, 1/2 sec, 1/10 sec, and 1/20 sec. The target 195

OCR for page 145
series consists of 18 targets, such as tanks, missiles, and anti- aircraft guns. Two film sequences comprise the total film pack- age, with the first film serving as the learning film. In the learning film, 14 views of each target, ranging from high-contrast photographs to low-contrast photographs, are presented in both slow and rapid order. The second film consists of an "eyeball training phase" in which 324 photographs of the 18 targets are presented to the subjects in random order in the high-speed mode only. Two groups of TAG line fighter pilots make up the sample. There are 20 pilots in each group, with one group re- ceiving conventional recognition training, as currently used in the field, and the second group receiving the special training outlined above. After completion of the training phase, both groups will be tested on 1-sec presentations of 72 views of the 18 targets. Following this, the pilots in each group will make target acquisition-recognition passes on four single targets placed on the range, perpendicular to the flight path. These passes will be flown at an airspeed of 550 knots at a 500-ft altitude. The two groups will be compared in regard to accu- rate target identification, distance at target acquisition, distance at target recognition, time lag between target acquisition and rec- ognition, and recall of details of the mission at time of debriefing. Such a technique appears to have a great deal to offer, not only in basic and specific programs of target recognition train- ing, but also in presenting to pilots, in a highly effective manner, other types of information such as foreign technical material, technical order data, low-level navigational routes with photo- graphs of checkpoints, and new reconnaissance data. All these data could be made available to the combat pilot and presented on a teaching machine for maximum effectiveness. Further, teaching machines are small and reliable, possessing great versatility, and film strips can be made up in a matter of days, hence, giving a reasonable reaction time for both large- and small-scale missions. SUMMATION The effort of the Deputy for Effectiveness Test in the area of low-altitude flight testing has been, to date, one of investigating various aspects of target acquisition and recognition, including evaluation of the techniques designed to improve the perceptual efficiency of pilots, and thus improve their capability for target 196

OCR for page 145
acquisition and recognition. Further tests are being established to evaluate target acquisition and recognition at various altitudes and speeds using fixed and mobile targets of varying contrasts. Future test efforts will involve search, acquisition, and recogni- tion of targets with live ordnance delivered against them. It is expected that these tests will be accomplished in 1964. While it is obvious that this total effort encompasses but a few aspects of the over-all low-altitude, high-speed flight prob- lem, the basic groundwork for solution of the problem is being carried out. It must be remembered that this effort was born of an operational need by a using command. This need must be satisfied in order to provide the necessary assistance to com- manders in the field, who are faced with the complex decisions of warfare, and who must have facts at their disposal to make their day-to-day "trade-off" between survival and effectiveness. REFERENCES Jones, R. L., & Lindsey, J. F. Human factors aspects of low-altitude flight: a sample of fighter pilot attitudes and altitude estimates. Eglin AFB: Air Prov. Ground Centr Doc., 1964, No. PGN 64-1 (APGC Proj. 9069Z). Joska, J. S. The effect of aircraft speed on low-altitude acquisition of ground targets. Eglin AFB: Air Prov. Ground Centr Doc., 1963, No. PGT 63-1. (APGC Proj. 5967Z2) 197

OCR for page 145

OCR for page 145

OCR for page 145

OCR for page 145

OCR for page 145

OCR for page 145
*- .„

OCR for page 145

OCR for page 145
"I. —