First published online December 28, 2007
Journal of Experimental Biology 211, 258-266 (2008)
Published by The Company of Biologists 2008
doi: 10.1242/jeb.012625
New experimental approaches to the biology of flight control systems
Graham K. Taylor1,*,
Marko Bacic2,
Richard J. Bomphrey1,
Anna C. Carruthers1,
James Gillies1,
Simon M. Walker1 and
Adrian L. R. Thomas1
1 Department of Zoology, Oxford University, South Parks Road, Oxford, OX1 3PS,
UK
2 Department of Engineering Science, Oxford University, Parks Road, Oxford, OX1
3PJ, UK
View larger version (28K):
[in this window]
[in a new window]
|
Fig. 1. Schematic diagram of the virtual-reality flight simulator. External
supports, etc., are not shown. The insect is tethered at the centre of the
acrylic sphere, which is suspended in a gasket. Two customized data projectors
project image sequences onto the entire surface of the sphere, which is
painted as a back-projection surface. The light path is folded using mirrors
to minimize the size of the apparatus and maximize resolution. The insect is
mounted on a six-component force–moment balance on the end of a movable
sting. In the configuration shown here, the sting is moved in an oscillatory
coning motion by a brushless motor, with two further adjustable axes providing
static adjustment of sting orientation to adjust the phasing of roll, pitch
and yaw. The insect sits at the open mouth of a transparent wind tunnel
mounted inside the sphere. The apparatus enables independent stimulation of
each of the sensory modalities involved in insect flight control.
|
|
View larger version (11K):
[in this window]
[in a new window]
|
Fig. 2. Photograph (left) of a 3D panoramic scene projected onto the surface of the
virtual-reality flight simulator. The scene shows the Oxford skyline with
grayscale `clouds' above for contrast and a checkerboard below. In this
simulated environment, the checkerboard is scrolled to simulate ventral
translational optic flow, and the whole scene is rotated to simulate
rotational optic flow. Note the mirror directly beneath the sphere; projection
continues behind the visible supports. The projected scene is generated in 3D
animation software using an explicit CAD-based representation of the flight
simulator. The CAD-generated image (right) shows the same exterior view; the
projected images are simply those that would be seen from the viewpoints of
the projectors.
|
|
View larger version (151K):
[in this window]
[in a new window]
|
Fig. 4. Inertial measurements from a steppe eagle in soaring flight. The graph
plots total measured acceleration against time: all three components of
acceleration, angular velocity and orientation are logged by the inertial
measurement unit, but are not shown. The thumbnails show synchronized frames
from a hand-held camcorder (upper row) to provide context, and from a
rearward-facing onboard camera (lower row) to confirm that the instrumentation
remains steady throughout. Dashed lines denote the correspondence of the graph
with the numbered frames. Note how the circled tan-coloured rump contour
feathers remain steady (position of circle identical between images),
indicating that the instrumentation is static with respect to the body. The
visible transients therefore denote real accelerations of the bird, and are
presumably excited by gusts, etc., as the bird is not actively manoeuvring in
this sequence. The downy white feathers that are visible on either side of the
circled contour feather are blowing freely in the wind, so provide no
information on the position of the instrumentation with respect to the
body.
|
|
View larger version (43K):
[in this window]
[in a new window]
|
Fig. 5. External photogrammetric measurement of the wing kinematics of a
free-flying steppe eagle Aquila nipalensis coming in to perch on its
handler's arm. Left panel shows one of a stereo pair of images taken at 500
frames s–1. Right panel shows a calibrated reconstruction of
the lower surface of the left wing based on stereo-matching of natural
features of the plumage. Black points on the wing denote measurements,
connected by straight lines to assist in visualizing the wing topography; the
colour map denotes the local geometric angle of attack of the interpolated
wing surface with respect to the horizontal. The isolated black points denote
reference measurements on the head and tail, indicating the longitudinal axis
of the bird. Note that whereas the angle of attack and camber of the proximal
section of the wing is relatively consistent in a spanwise direction, the
distal portion of the wing is set at a much greater angle of attack. This
reflects the angle of attack of the interpolated surface and does not take
account of the local twist of the primary feathers, which will be measured in
future work. An animation of this perching sequence is available (Movie 1 in
the supplementary material).
|
|
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati 
Twitter What's this?
© The Company of Biologists Ltd 2008
) is extracted from the angle of the
trailing edge, as shown by the construction lines on the image. Tail pitch
angle (
) is extracted by measuring the deviation of the trailing edge
from its average position perpendicular to the line AB at the point A, and
making use of the known distance of the camera to the base of the tail. Tail
spread angle is not shown, but can also be determined from these data, giving
3 measurable kinematic degrees of freedom for the tail.