KC-135 POSTFLIGHT REPORT

Figure 1: Test Subjects Daniel Maynes-Aminzade and Kate Williams

TITLE: IPAGE: Improved Productivity in Altered Gravity Environments

FLIGHT DATES: March 8th and March 9th, 2000

PRINCIPAL INVESTIGATORS: Carnegie Mellon University

Advisor: Dr. Scott Hudson

Journalist: Dr. Randy Pausch

Students: William Emmer

Daniel Maynes-Aminzade

Kathleen Williams

Peter Pak-Shum Yeung

GOAL:

The overall goal of our project was to create and test a Virtual Reality preflight trainer for first time flyers aboard the KC-135 aircraft. Our system (IPAGE: Improved Productivity in Altered Gravity Environments) utilized unique technology available at Carnegie Mellon University, and is an ongoing project within CMU’s Human Computer Interaction Institute. The system, by fostering interaction with a “Virtual KC-135,” aimed to prepare first-time KC-135 flyers for the effects of altered gravity before their flight, so as to improve their performance and comfort once aboard the aircraft itself.

OBJECTIVES:

Our experiment’s objectives were as follows:

(1) To test how our Virtual Reality preflight trainer affected the productivity of our team of first-time KC-135 flyers,

(2) To note what improvements we could make to our existing training system, and

(3) To examine how Virtual Reality programmers might better develop rapid simulations of unique and unfamiliar environments.

INTRODUCTION / PROJECT ABSTRACT:

Carnegie Mellon University's Human Computer Interaction Institute has created a unique Virtual Reality system that provides superior simulation of environments as well as an interface that allows for rapid creation of "virtual worlds." It is our belief that this system has exceptional potential concerning the amelioration of current preflight simulation and adaptation methods. There are many aspects that remain unexplored concerning Virtual Reality training methods, as well as concerning the neurovestibular effects of altered-gravity environments. We view our KC-135 experiment as a small step in Virtual Reality research, with the hope that our findings (as well as the VR system itself) may be of interest and of use to current researchers in preflight training and simulation.

METHODS AND MATERIALS:

For preflight training, we utilized a Virtual Reality system developed in conjunction with Carnegie Mellon’s Stage Three labs. A “mission toolkit” was built for the in-flight portion of our experiment, and included such rudimentary components as a target board, a shape board, a pattern matcher, and a tethered clipboard. These were used to quantify the in-flight coordination of each subject.

SUBJECTS:

The subjects of this experiment were the undergraduate members of the project team: William Emmer, Daniel Maynes-Aminzade, Kate Williams, and Peter Yeung. For the experiment, Emmer and Maynes-Aminzade trained for 4 hours in the altered-gravity “Virtual KC-135” preflight training environment, while Williams and Yeung trained for 4 hours in the “Virtual KC-135” with no altered-gravity effects (constant 1g).

INSTRUMENTS:

The “Virtual KC-135” preflight training system was run on a standard off-the-shelf computer (Pentium processor), using Carnegie Mellon’s Alice software. Alice is a 3D Interactive Graphics Programming Environment for Windows 95/98/NT built by the Stage 3 Research Group at Carnegie Mellon University. The Alice project is a public service to the wider computing and artistic communities; its hope is to make it easy for novices to develop interesting 3D environments and to explore the new medium of interactive 3D graphics. The current version of Alice authoring tool is free to everyone and runs on computers that are commonly available for reasonable prices. Worlds created in Alice can be viewed and interacted with inside of a standard web browser once the Alice plug-in has been installed.

PROCEDURE:

Our procedure can be divided into two sections:

(1) Preflight VR simulation training, and

(2) In-flight coordination testing.

For the preflight training section, half the team members trained in the altered-gravity Virtual KC-135 environment, while the other half trained in a Virtual KC-135 that used regular 1g conditions. Training entailed practicing a rudimentary coordination function (trajectory test - detailed below) within the VR world. All training and testing functions were chosen for their common proficiency levels among test subjects.

For the coordination testing aboard the KC-135, we quantitatively measured each subject’s effectiveness in completing simple tasks under altered gravity conditions. For each flight, one subject having trained in the altered-gravity VR flew with a subject having trained in the 1-g VR. For the in-flight test, subjects took alternating turns performing the task and timing/evaluating the other flyer performing the same task. Our data was recorded via a stopwatch and data acquisition sheets.

Trajectory

The trajectory test was designed to develop (and, while in-flight, measure) a subject’s ability to adapt his/her interaction with a target board in an altered-gravity environment. A picture of this test within the Virtual KC-135 is shown below.

Figure 3: Virtual KC-135 preflight trainer with Trajectory Test

The in-flight component of the trajectory test evaluated the accuracy with which test subjects threw a ball at a target in micro-gravity. The target board was placed on the side of the mission toolkit, and consisted of a number of concentric circles of increasing diameter and numbered in such a manner to quantify the thrower’s closeness to the “bull’s eye”. The test subject was positioned at a specified distance (approx. 6 feet) from the target and then threw the ball, while the other flyer recorded the “score” of the throw (“misses” were also recorded). The possible scores on the target board ranged from 100 points (bull’s eye) to 10 points (outermost ring). The target wall and ball were covered in a Velcro and felt configuration such that the Velcro on the ball stuck to its target upon arrival. This assisted in accurate data recording.

Figure 4: In-flight Trajectory Test: Test Subjects Peter Yeung and William Emmer

RESULTS:

The following chart represents the averaged score per hit and average number of misses for the two sets of test subjects (altered-g VR training and regular-g VR training). The blue data represents the average results from the two subjects who had trained on the altered-gravity “Virtual KC-135”, while the red represents the two who had trained in the regular-gravity “Virtual KC-135.” The total number of attempted throws per test subject for Flight One (Maynes-Aminzade and Williams) was 30; the total number of attempted throws each for Flight Two (Yeung and Emmer) was 35.

Figure 5: In-Flight Performance: Altered-G Training vs. 1-G Training

REVIEW OF TEST RESULTS:

In terms of our recorded results, we cannot with good conscience come to a conclusion concerning the quantitative difference in performance between the two test subject pools. The number of test subjects (4) was too low to determine whether or not the differences in accuracy and precision are attributed to the differences in training regimen.

Qualitatively speaking, however, the test subjects all expressed the effectiveness of their training program in making them feel more prepared for their flight aboard the KC-135. In addition, all four subjects noted that the over four hours of immersion in the Virtual Reality cabin made them feel more comfortable and confident within the actual plane. Out of the four test subjects, only one displayed any signs of motion sickness; it is our suspicion that this perceived amount of preflight preparation perhaps allowed the subjects to resist motion sickness more effectively than the general population of first-time flyers.

CONCLUSION:

Due to the small test subject pool, no conclusive evidence was found to support or refute the hypothesis that the altered-g Virtual KC-135 enhanced the accuracy or precision of its users. However, the overall perceived level of comfort and preparedness – along with these factors’ correlation to motion sickness – supports the assertion that Virtual Reality preflight training could still be an effective tool for first-time KC-135 flyers.

The first-time flyer test subjects, having been also the primary developers of the Virtual KC-135 preflight training system, found that flying aboard the KC-135 presented many different challenges from what they had factored into the VR trainer. For future Virtual KC-135 preflight trainers, issues of timing, presence of self in a zero-g environment, and feedback for moving your head in 2g are only some of the suggested additions to the regimen.

REFERENCES:

http://www.etc.cmu.edu/Projects/KC-135

http://www.tgsc.utexas.edu/floatn/march2000/teams

Dr. Deborah Harm (JSC-SD)