The fly
flight arena was designed (not by me!) to explore the connections between
the different sensory modalities that fruit flies use to control their
flight. The fly is glued to a metal post mounted in the center of a
cylindrical arena. The walls of this cylinder are made out of 11,340
LEDs which are controlled in real time by a computer. (Flies have poor
spatial resolution, estimated at 5°, but very fast temporal resolution
- around 200 Hz. Human vision has spatial resolution of about 1/30th
degree and temporal resolution around 20 Hz.)
This arena is mounted in a 3 degree-of-freedom gimbal. The gimbal is
designed to rotate the arena (with the fly inside) around all 3 axes
(yaw, pitch, roll) independently. This is also controlled in real time
by a computer. The rig was designed to achieve velocities up to 2000°/sec
(333 rpm) on each axis, with accelerations up to 20000°/sec_, coinciding
with estimates of the forces actually acting on flies in flight. The
fly senses rotational movements of its body using its halteres, modified
hindwings that act as onboard gyroscopes.
This entire apparatus is mounted on a Newport table with custom 80/20
housing. Control and data acquisition are done by three PCs using a
wide variety of software and hardware, some of which was custom-designed
for this rig. Graphical user interfaces allow hands-off operation and
precise control for experiments.
The fly is illuminated by an infrared diode, and the shadow cast by
its wings on a photosensor is used to determine its intended flight
dynamics. This information is used by the computers to feed back onto
the visual and rotational environment in real time. In this way, the
fly can control its own flight experience, much like a virtual reality-type
arcade game.
The purpose of my experiments is to understand how the fly integrates
information from visual and rotational stimuli, and the relative importance
of each sensory modality in determining the fly's behavior. Using this
arena, I can stimulate the flies in their visual and mechanosensory
(rotation-sensitive) senses independently, and monitor the resulting
behavior to understand how each sense contributes to behavior. Fruit
flies are much simpler than humans or even other insects, so their reactions
to these stimuli are more predictable and allow more insight into the
underlying biological functions.
Figure
1. The rig, table, housing, wiring
Figure
2. The 3 computers required to run the rig
Figure
3. Some of the hardware for control and data acquisition
Figure
4. A closer view of the 3 degree-of-freedom gimbal with the flight
arena mounted inside
Figure
5. For scale...
Figure
6. A closer view of the flight arena
Figure
7. Inside the arena; the fly gets attached to the metal post projecting
from right to left across the bottom of the image
A movie of this rig
in action (although not at maximum speed) can be found at http://vision.caltech.edu/~jbender/movies/rocnroll.avi.
(This AVI movie is about 145 MB.)
