VISUAL-FIELD INFORMATION IN LOW-ALTITUDE VISUAL FLIGHT BY LINE-OF-SIGHT SLAVED HELMET-MOUNTED DISPLAYS

Citation
Aj. Grunwald et S. Kohn, VISUAL-FIELD INFORMATION IN LOW-ALTITUDE VISUAL FLIGHT BY LINE-OF-SIGHT SLAVED HELMET-MOUNTED DISPLAYS, IEEE transactions on systems, man, and cybernetics, 24(1), 1994, pp. 120-134
Citations number
17
Categorie Soggetti
Controlo Theory & Cybernetics","Computer Science Cybernetics","Engineering, Eletrical & Electronic
ISSN journal
00189472
Volume
24
Issue
1
Year of publication
1994
Pages
120 - 134
Database
ISI
SICI code
0018-9472(1994)24:1<120:VIILVF>2.0.ZU;2-A
Abstract
The pilot's ability to derive control-oriented visual field informatio n from teleoperated helmet-mounted displays in nap-of-the earth flight is investigated in this paper. The visual field with these types of d isplays, commonly used in Apache and Cobra helicopter night operations , originates from a relatively narrow field-of-view forward looking in frared radiation (FLIR) camera, gimbal-mounted at the nose of the airc raft and slaved to the pilot's line of sight, providing a wide-angle h eld of regard. Pilots have encountered considerable difficulties in co ntrolling the aircraft by these devices. The experimental simulator re sults presented here indicate that part of these difficulties can be a ttributed both to the narrow camera field of view and to head/camera s laving system phase lags and errors. In the presence of voluntary head rotation, these shortcomings are shown to impair the control-oriented visual field information vital in vehicular control, such as the perc eption of the anticipated flight path or the vehicle yaw rate. Since t he pilot will tend to minimize head rotation in the presence of slavin g system imperfections, the full wide-angle field of regard of the lin e-of-sight slaved helmet-mounted display is not always fully utilized. The findings in this paper are valid for a general class of head-slav ed displays which are used in teleoperation and virtual environments a nd in which correct self-motion estimation is an essential part of the operator task.