Use of a Head-Worn Display for Approach and Landing in a Transport Category Aircraft: Does Monocular Viewing Impact Pilot Performance and Workload?

Abstract

When manually flying an approach and landing, pilots can use a Head-Up Display (HUD), binocular Head-Worn Display (HWD), or monocular HWD to view flight symbology that is superimposed onto their view of the runway environment. There are design differences across these displays that could impact pilots’ performance and workload while flying. Based on these concerns, an experiment was conducted where 18 Airline Transport Pilot Captains flew approach, landing, and rollout scenarios with varying visibility levels in a Level D-equivalent Boeing 737 flight simulator. Pilots manually flew each scenario while using flight symbology presented on a HUD, binocular HWD, or monocular HWD. Pilots rated their workload during each scenario using the NASA-TLX. While flightpath tracking and energy management performance was not impacted by display type, pilot workload was highest when flying with a Monocular HWD and lowest when flying with a HUD. The operational takeaways and theoretical implications are discussed.

Objectives

The approach and landing phase of flight is among the most safety-critical operations for pilots of fixed-wing aircraft. It is associated with increased pilot workload relative to other phases of flight, particularly when weather conditions restrict pilots’ ability to see the runway during a landing. The Boeing Aircraft Co. (2023) reported that, despite representing 4% of the total flight time of a typical flight, 46% (15 of 32) of fatal accidents involving transport category aircraft between 2013 and 2022 occurred during approach and landing.

To mitigate these risks, transport category aircraft can be equipped with a Head-Up Display (HUD), which superimposes flight instruments onto the pilot’s view outside the aircraft using a transparent display. These instruments are presented at a focal distance of optical infinity, enabling the pilot to monitor flight path and airspeed information during an approach and landing while maintaining visual contact with the runway (Weintraub & Endsing, 1992). Recently, manufacturers have begun developing Head-Worn Displays (HWDs) for transport category aircraft (Federal Aviation Administration Flight Technologies and Procedures Division, 2022). Like the HUD, the HWD enables the pilot to view aircraft instruments while maintaining visual contact with the runway.

There are fundamental differences between the HUD and the HWD that warrant a comparison between these technologies in terms of their impact on pilot performance and workload. Some HWDs may be designed so the symbology appears at a distance less than optical infinity. This may increase the degree to which the pilot’s eyes adjust focal length and vergence angle when switching attention between the symbology and the runway, degrading their ability to divide attention between HWD symbology and the runway environment (Condino et al., 2019). The use of a near-to-eye display may also constrain the pilot’s view of other information in the flight deck environment.

HWDs can display symbology to both eyes (i.e., binocular) or to a single eye (i.e., monocular) depending on its physical configuration. There is evidence to suggest that monocular presentation of visual information may negatively impact performance on visual tasks, suggesting that there may be negative impacts on pilot performance during a low-visibility approach and landing when pilots use a monocular HWD (Winterbottom et al., 2006). Using symbology that is presented to a single eye may also be more demanding than using the same symbology presented to both eyes. This could elevate workload compared to binocular presentation of that same information.

Based on these concerns, we conducted an experiment to determine whether pilot performance and workload during a low-visibility approach and landing differed depending on whether the pilot used a HUD, binocular HWD, or monocular HWD to view flight symbology. The experiment addressed the following questions: (1) does a pilot’s ability to follow flightpath guidance, maintain target airspeed, and land on the runway centerline in low-visibility conditions differ as a function of Display Type (i.e., HUD, binocular HWD, or monocular HWD)? (2) Does a pilot’s ability to land on the runway centerline in low-visibility conditions differ as a function of runway visual range (RVR) and Display Type? (3) Does pilot workload during an approach and landing differ as a function of Display Type and RVR?

Approach

The findings reported here are part of a larger, ongoing research effort. To date, nine Airline Transport Pilot (ATP) crews, consisting of current HUD-qualified Boeing 737 Type Rated airline Captains, flew instrument landing system (ILS) approach and landing scenarios in a Boeing 737 Level D-equivalent flight simulator under a variety of visibility conditions using a HUD, Binocular HWD, or Monocular HWD. To elevate workload, all scenarios involved quartering headwinds with gusts. For HUD scenarios, pilots used a Collins Aerospace HGS-6700 HUD. For HWD scenarios, pilots used a Microsoft HoloLens 2 that presented the HUD symbology. A monocular HWD condition was achieved by disabling the non-dominant eye image source on the HoloLens 2. Overcast clouds and ground-level fog restricted RVR (i.e., the longitudinal distance the pilot is able to see down the runway). Simulator motion was disabled to prevent interference with HWD head tracking.

Each crew flew 18 approach and landing scenarios using either a HUD, binocular HWD, or monocular HWD with either 4,800, 1,200, or 600 ft of RVR. All scenarios involved normal operations. Pilots alternated Pilot Flying and Pilot Monitoring roles throughout the study so that each pilot flew nine scenarios as Pilot Flying. Pilot performance was evaluated in terms of root mean square (RMS) deviation from the flight path of the approach, RMS deviation from the target airspeed of the approach, and lateral deviation from the runway centerline at landing. At the end of each scenario, pilots rated their workload using the NASA-Task Load Index (TLX; Hart & Staveland, 1988).

Findings

Analyses of data collected thus far revealed that there was not a significant effect of Display Type on deviation from flight path in instrument segment (p = .337; see Supplemental Figure 1). There was also no significant effect of Display Type on deviation from target airspeed during approach (p = .186; see Supplemental Figure 2). There was a significant effect of RVR on lateral deviation from runway centerline at landing (F(2, 150) = 5.17, p = .007, η_p² = .064). Trends among the means indicate that lateral deviation from runway centerline at landing is highest with 600 ft. RVR, second highest with 1,200 ft. RVR, and lowest with 4,800 ft. RVR (see Supplemental Figure 3). There was not a significant effect of Display Type on lateral deviation from runway centerline at landing (p = .329).

There was a significant effect of Display Type on NASA-TLX total weighted score (F(2, 150) = 18.65, p < .001, η_p² = .199). Trends among the means indicate that NASA-TLX total weighted score was elevated during scenarios with a Monocular HWD compared to scenarios with a Binocular HWD or HUD; NASA-TLX total weighted score was elevated during scenarios with a Binocular HWD compared to scenarios with a HUD (see Supplemental Figure 4). There was a significant effect of RVR on NASA-TLX total weighted score (F(2, 150) = 3.39, p = .036, η_p² = .043). Trends among the means indicate that NASA-TLX total weighted score was highest during 600 ft RVR scenarios, second highest during 1,200 ft RVR scenarios, and lowest during 4,800 ft RVR scenarios.

Takeaways

While there is evidence that performance on visual tasks tends to degrade when visual information is presented to a single eye, it does not appear that this effect translates to impacts on a pilot’s performance during a low-visibility approach and landing with a monocular HWD relative to performing the same operation with a binocular HWD. This also appears to be the case when comparing the HUD to the HWD, where pilot performance with a HUD did not differ from performance with a binocular HWD.

Conversely, workload was elevated when using a monocular HWD compared to when using a binocular HWD, indicating that using flight symbology that is presented to only one eye during an approach and landing places more demands on the pilot’s attention than when symbology is presented to both eyes. Elevated workload when using a binocular HWD compared to when using a HUD indicates that the characteristics of the HWD used in this study increase the demands of the approach and landing relative to using a HUD.

These findings suggest that the existing level of pilot performance when using a HUD for a low-visibility approach and landing may be preserved when pilots use a monocular or binocular HWD. Future research should determine whether this applies when an abnormal event occurs, such as a runway incursion or aircraft malfunction. Elevated workload when using a monocular or binocular HWD may increase the risk of attentional tunneling; in turn, this may negatively impact the pilot’s ability to detect abnormal events (Wickens & Alexander, 2009).

Supplemental Material

sj-pdf-1-pro-10.1177_10711813241280042 – Supplemental material for Use of a Head-Worn Display for Approach and Landing in a Transport Category Aircraft: Does Monocular Viewing Impact Pilot Performance and Workload?

Supplemental material, sj-pdf-1-pro-10.1177_10711813241280042 for Use of a Head-Worn Display for Approach and Landing in a Transport Category Aircraft: Does Monocular Viewing Impact Pilot Performance and Workload? by David C. Newton, Eric T. Greenlee, Theodore C. Mofle, Farzaneh Shahini, Rebecca DiDomenica and Inchul Choi in Proceedings of the Human Factors and Ergonomics Society Annual Meeting