Motion Fidelity Effects With Mixed-Reality

The aim of this study is to determine the effects of simulator motion fidelity on pilot performance in two demanding piloting tasks while wearing a mixed-reality headset for OTW visuals. Pilots will be exposed to mixed-reality visuals meaning that the virtual OTW visual scene will be overlayed on an image of the real simulator cockpit and head down displays (HDD). Pilots will fly two tasks in this experiment: 1) a roll-lateral sidestep maneuver, and 2) a precision landing.

In task 1, the roll-lateral sidestep maneuver will be from one landing pad to a second landing pad of a vertiport on the Fifth \& Mission Parking Garage in San Francisco. The vertiport is in a San Francisco visual database surrounded by buildings with high-resolution textures. Pilots are instructed to stay at a constant altitude. A positioning sight will be placed in the visual scene at each landing pad to allow pilots to determine their positioning error precisely. Pilots will be provided with lateral position and time criteria for desired and adequate performance. Pilots will be instructed to press an event marker on their control inceptor when they acquire the station-keeping point at the landing pad. Turbulence will be used to make the task more difficult. This task will be similar to the roll-lateral experiment described by Schroeder that was also performed in the VMS.

In task 2, pilots will perform a landing at the same vertiport in San Francisco that is used in task 1. The task will start from level flight from a position that allows for a standard approach and landing. Landing position and time criteria for desired and adequate performance will be communicated to the pilots. Pilots will press an event marker on the control inceptor when they acquire the desired reference point on the landing pad. Turbulence will be introduced to make the task more difficult.

Pilots will receive the informed consent form and a Motion Sickness Susceptibility Questionnaire (MSSQ via email first. They will be asked to fill out the MSSQ at home and return it to the PI before being scheduled and coming to the VMS. On the day of the experiment, participants will first receive a briefing and walkaround of the facility before providing their informed consent (60 minutes). Next, they will take a seat in the simulator cab where seat and headset adjustments will be made first to make sure the experiment can be performed comfortably. This is followed by the first data-collection session in which one of the tasks is flown (45 minutes). Both tasks will be flown under three motion fidelity conditions described in Section 8.3 below. Three repetitions of each motion condition will be flown for both tasks, after completing three training runs, for a total of 12 runs per task. A 30-minute break is scheduled after the first data-collection session. The second session in which the other tasks is flown will commence after the break (45 minutes). Finally, pilots will be debriefed and can ask any remaining questions (30 minutes).

Pilot performance in each task will be determined by several outcome-based variables, including the error between desired and actual aircraft position and attitude at specified locations, the intensity of control inputs, and task completion time. Pilots will also complete a motion fidelity rating and a Simulator Sickness Questionnaire (SSQ) at the end of each run.

Hypothesis: Pilot performance and subjective motion ratings will be significantly different between the different simulator motion fidelity conditions.

Before data analysis, all data will be averaged over the three repetitions of each condition. A one-way repeated-measures Multivariate Analysis of Variance (MANOVA) will be used on the data from each task to determine statistically significant effects. If test assumptions are not met (e.g., homogeneity of covariance matrices), corrections or appropriate alternate tests will be used. Any reporting of the statistical results will include the F statistic (or equivalent for alternate tests), degrees of freedom, probability values, and effect sizes.

Based on previous studies investigating the effects of motion fidelity on pilot performance, it is expected that pilot performance will be lower without simulator motion or with small-hexapod motion compared to one-to-one vehicle motion. Control intensity will show the opposite trend; that is, will be higher without simulator motion and will decrease with higher motion fidelity.