Masters Theses

Date of Award

8-1997

Degree Type

Thesis

Degree Name

Master of Science

Major

Aviation Systems

Major Professor

F. Collins

Committee Members

Lewis, Kimberlin

Abstract

This thesis presents the results of a developmental test of a prototype G-Induced Loss of Consciousness (G-LOC) Detection System installed on a T-38A aircraft. The purpose of the test was to develop a functioning G-LOC Detection System that was properly integrated into the airborne environment. The test was carried out at the Air Force Flight Test Center, Edwards AFB, California from February through May 1991. Eight test sorties were flown for a cumulative total of 6.7 flight test hours. The program was requested by NASA-Ames-Dryden, and Eidetics International, Incorporated was the manufacturer of the G-LOC Detection System.

The G-LOC phenomenon and the physiological requirements and constraints associated with successful implementation of a non-obtrusive detection system are expored. A careful review of the state-of-the-art technologies in GLOC detection is also conducted.

The G-LOC Detection System evaluated was a simple aircraft sensor-based system designed to detect G-LOC using T-38A Data Acquisition System (DAS) inputs. The G-LOC system continuously calculated an instantaneous probability of pilot G-LOC based on the aircraft load factor time history. The system then monitored various DAS inputs including stick movement and force, throttle and rudder pedal movement, stick grip, and a cockpit activity button to determine if there was in fact cockpit activity. If there was no cockpit activity for 3 seconds and the probability of pilot G-LOC was greater than 50 percent, the system would detect a G-LOC condition which would be reflected with a G-LOC detection light indication in the T-38A rear cockpit.

The overall objective of this program was to develop the prototype detection system so that it accurately recognized whether or not simulated G-LOC had occurred during operationally representative maneuvering in a fighter type aircraft. After five sorties this objective was accomplished, and the G-LOC Detection System was able to recognize when G-LOC was and was not simulated.

The G-LOC Detection System met its basic design objective. The system operated precisely as designed on aU test maneuvers following the first 5 developmental sorties. The implementation of this type of G-LOC Detection System is therefore considered feasible. Consequently, a system designed to unobtrusively detect G-LOC using aircraft data shows promise. Although the system functioned exactly as designed, it contained logic flaws which made it inadequate for military use. The logic flaws included the modeling of pilot stick grip which caused detection failures; the modeling of conscious pilot actions such as afterburner selection/de-selection and switch activations which caused false alarms; and an inaccurate GLOC probability build-up function. These findings were based on test team experience and were supported with Air Force accepted physiological data.

Finally, improvement proposals (algorithm logic and sensor inputs) are made to better implement this type of detection system in more advanced fighter aircraft where substantially more aircraft and pilot status information is readily available for processing.

Download
Files over 3MB may be slow to open. For best results, right-click and select "save as..."

Share

COinS