Helmet-mounted display (HMD) designs have faced persistent
head-supported mass and center of mass (CM) problems,
especially HMD designs like night vision goggles (NVG) that utilize image intensification (I2) sensors mounted forward
in front of the user's eyes. Relocating I2 sensors from the front to the sides of the helmet, at or below the transverse plane
through the user's head CM, can resolve most of the CM problems. However, the resulting increase in the separation
between the two I2 channels effectively increases the user's interpupillary distance (IPD). This HMD design is referred to
as a hyperstero design and introduces the phenomenon of hyperstereopsis, a type of visual distortion where stereoscopic
depth perception is exaggerated, particularly at distances under 200 feet (~60 meters). The presence of hyperstereopsis
has been a concern regarding implementation of hyperstereo HMDs for rotary-wing aircraft. To address this concern, a
flight study was conducted to assess the impact of hyperstereopsis on aircraft handling proficiency and pilot acceptance.
Three rated aviators with differing levels of I2 and hyperstereo HMD experience conducted a series of flights that
concentrated on low-level maneuvers over a two-week period. Initial and final flights were flown with a standard issue
I2 device and a production hyperstereo design HMD. Interim flights were flown only with the hyperstereo HMD. Two
aviators accumulated 8 hours of flight time with the hyperstereo HMD, while the third accumulated 6.9 hours. This paper
presents data collected via written questionnaires completed by the aviators during the post-flight debriefings. These data
are compared to questionnaire data from a previous flight investigation in which aviators in a copilot capacity, hands not
on the flight controls, accumulated 8 flight hours of flight time using a hyperstereo HMD.
Modern helmet-mounted night vision devices, such as the Thales TopOwl<sup>TM</sup> helmet, project imagery from intensifiers
mounted on the sides of the helmet onto the helmet visor. This increased effective inter-ocular separation distorts several
cues to depth and distance that are grouped under the term "hyperstereopsis". Stereoscopic depth perception, at near to
moderate distances (several hundred metres), is subject to magnification of binocular disparities. Absolute distance
perception at near distances (a few metres) is affected by increased "differential perspective" as well as an increased
requirement for convergence of the eyes to achieve binocular fixation. These distortions result in visual illusions such as
the "bowl effect" where the ground appears to rise up near the observer. Previous reports have indicated that pilots can
adapt to these distortions after several hours of exposure. The present study was concerned with both the time course and
the mechanisms involved in this adaptation. Three test pilots flew five sorties with a hyperstereo night vision device.
Initially, pilots reported that they were compensating for the effects of hyperstereopsis, but on the third and subsequent
sorties all reported perceptual adaptation, that is, a reduction in illusory perception. Given that this adaptation was the
result of intermittent exposure, and did not produce visual aftereffects, it was not due to the recalibration of the
relationship between binocular cues and depth/distance. A more likely explanation of the observed visual adaptation is
that it results from a discounting of distorted binocular cues in favour of veridical monocular cues, such as familiar size,
motion parallax and linear perspective.
A flight study was conducted to assess the impact of hyperstereopsis on helicopter handling proficiency, workload and
pilot acceptance. Three pilots with varying levels of night vision goggle and hyperstereo helmet-mounted display
experience participated in the test. The pilots carried out a series of flights consisting of low-level maneuvers over a
period of two weeks. Four of the test maneuvers, The turn around the tail, the hard surface landing, the hover height
estimation and the tree-line following were analysed in detail. At the end of the testing period, no significant difference
was observed in the performance data, between maneuvers performed with the TopOwl helmet and maneuvers
performed with the standard night vision goggle. This study addressed only the image intensification display aspects of
the TopOwl helmet system. The tests did not assess the added benefits of overlaid symbology or head slaved infrared
camera imagery. These capabilities need to be taken into account when assessing the overall usefulness of the TopOwl
system. Even so, this test showed that pilots can utilize the image intensification imagery displayed on the TopOwl to
perform benign night flying tasks to an equivalent level as pilots using ANVIS. The study should be extended to
investigate more dynamic and aggressive low level flying, slope landings and ship deck landings. While there may be
concerns regarding the effect of hyperstereopsis on piloting, this initial study suggests that pilots can either adapt or
compensate for hyperstereo effects with sufficient exposure and training. Further analysis and testing is required to
determine the extent of training required.
Providing both I<sup>2</sup> (image intensified) and FLIR (forward looking infrared) images on a helmet-mounted
display (HMD) requires perceptual design tradeoffs. Primary considerations center on the number, type,
and placement of sensors. Perceptual drivers for these tradeoffs are derived from monocular versus
biocular/binocular displays and offset of the sensors from the design eye. These conditions can create
binocular rivalry, perceptual perspective distortion or hyperstereopsis, a binocular perceptual distortion that
occurs when the sensors are positioned further apart than the interpupillary distance (IPD). Each of these
perceptual tradeoff considerations is discussed.
A number of currently proposed helmet-mounted display (HMD) designs relocate image intensification (I<sup>2</sup>) tubes to
the sides of the helmet. Such a design approach induces a visual condition referred to as hyperstereo vision (or
hyperstereopsis). This condition manifests itself to the user as an exaggerated sense of depth perception, causing
near- to mid-range objects to appear closer than they actually are. Hyperstereopsis is potentially a major concern for
helicopter operations that are conducted at low altitudes. As part of a limited flight study to investigate this
phenomenon, five rated U.S. Army aviators, as technical observers, wore a hyperstereo HMD during the conduct of
a series if 13 standard maneuvers. Two subject aviators acquired a total of eight hours and three aviators a single
hour of flight. Using a post-flight questionnaire, these aviators were asked to compare their visual experiences to
that of normal I2-aided flight. Depth perception at distances below 300 feet was identified as the greatest challenge.
The two 8-hour aviators reported a 5-8 hour "adaptation" period for most maneuvers.
This is the last in a series of studies designed to establish objective tools that can evaluate vision and cognitive visual performance with head-/helmet-mounted displays (HMDs) in a static environment. These tools will be applied to ongoing studies in an interactive, animated environment that will compare visual performance of biocular and monocular HMDs and evaluate image misalignment tolerance standards during long-term wear (4-6 hours) and during vibration. In this study, saccadic velocity significantly decreased following image misalignment, but not following a corresponding control period without misalignment, showing the measurement to be sensitive to image misalignment in a full-overlap biocular display. Pupil constriction latency significantly increased for both the experimental and control conditions, indicating that, although sensitive to using the optical device, this measurement was not specifically sensitive to a divergent horizontal misalignment of 3.1 milliradians (mrad). These results add saccadic velocity to the tools that we have already established. We also evaluated the visual performance of presbyopic subjects using a full-overlap display and found decreased performance with and without image misalignment. This has implications for helmet-mounted display use by older pilots that certainly warrants further investigation. In previous studies<sup>6, 7</sup> individuals with accommodative and vergence problems showed reduced performance when there was a small divergent horizontal image misalignment in a partial-overlap biocular display.
Partial-overlap biocular helmet-mounted display (HMD) design flexibility and cost are directly related to image misalignment standards. Currently suggested standards are based on highly variable data from a number of studies, most using subjective discomfort or diplopia measures. This study tested the suggested standard for divergent horizontal image misalignment in a partial-overlap biocular optical system by exercising vigilance performance during image misalignment. Also, pre- and post-image misalignment divergence, convergence and heterophoria measurements were taken. The results revealed that clinical visual diagnoses, associated with accommodation and vergence, were clearly related to vigilance task performance, showing a greater number of vigilance errors for subjects viewing misaligned displays. In-device post-image misalignment divergence recovery and convergence break-recovery were significantly decreased. This was not found for the no-offset controls.
Biocular helmet-mounted display (HMD) design flexibility and cost are directly related to image misalignment tolerance standards. Currently recommended tolerance levels are based on highly variable data from a number of studies. This paper presents progress of an ongoing study to evaluate optometric measures sensitive to misalignment in partial-overlap biocular optical systems like that proposed for the Comanche RAH-66 helicopter helmet integrated display sighting system (HIDSS). Horizontal divergent and relative vertical misalignments (offsets) of see-through biocular symbology viewed against a simulated daytime background were chosen for this study. Misalignments within and just beyond current tolerance recommendations were evaluated using pre, pre and post, and during measures of visual performance. Data were obtained from seven experimental and four control subjects. The diplopia responses from experimental and control subjects were essentially the same. However, accommodative facility showed a rate decrement following exposure to both types of misalignment. Horizontal heterophorias showed definite post-misalignment increases. Subject responses to questionnaires universally indicated increased adaptation to (ease with) visual tasks over the testing period.
To meet the goal of 24-hour, all-weather operation, U.S. Army aviation uses a number of imaging sensor systems on its aircraft. Imagery provided by these systems is presented on helmet-mounted displays (HMDs). Fielded systems include the Integrated Helmet Display Sighting System (IHADSS) used on the AH-64 Apache. Proposed future HMD systems such as the Helmet Integrated Display Sighting System (HIDSS) and the Microvision, Inc., Aircrew Integrated Helmet System (AIHS) scanning laser system are possible choices for the Army's RAH-66 Comanche helicopter. Ever present in current and future HMD systems is the incompatibility problem between the design-limited physical eye relief of the HMD and the need to provide for the integration of laser and nuclear, biological and chemical (NBC) protection, as well as the need to address the changing optical and vision requirements of the aging aviator. This paper defines the compatibility issue, reviews past efforts to solve this problem (e.g., contact lenses, NBC masks, optical inserts, etc.), and identifies emerging techniques (e.g., refractive surgery, adaptive optics, etc.) that require investigation.