Keeping large
structures stable in space
BY DAVID F. SALISBURY
So far, manned spacecraft
have been small, cramped capsules.
Much larger orbital
structures, like the international space station, are now
on the drawing boards. As these larger space habitats are
constructed, a new problem will become increasingly
important: stability.
If two sections of a space
structure begin moving to the beats of different
drummers, the structure could easily be severely damaged.
One solution to the
problem is to design in added rigidity, but that adds
weight, which can significantly increase cost.
An alternate, and
potentially more cost-effective, approach is to build in
a dynamic control system that actively controls errant
oscillations. E. Harrison Teague, a Stanford doctoral
student in aeronautics and astronautics, has developed
such a system, which employs signals from the Global
Positioning System, the Department of Defense's satellite
navigation system.
Using inexpensive GPS
receivers attached to different portions of a space
structure, the method can detect wayward motions with
centimeter-level precision and then automatically fire
thrusters to compensate for them. The system also can be
used to change the orientation of a flexible structure
with such accuracy that it moves almost as if it were
rigid.
Teague developed the GPS
system as part of his doctoral thesis, which he completed
in June. An article describing the work will appear in
the summer issue of the Navigation Journal, which
is still in press. His thesis advisers were aeronautics
and astronautics professors Jonathan How and Bradford
Parkinson.
Previous methods that
provided centimeter-level measurements of position and
attitude using GPS relied on the object in question being
a rigid body. Teague adapted these techniques to provide
the same level of precision with a flexible structure.
Next he had to identify
the shapes and frequencies of the various modes of
oscillation that could develop in such a structure.
Although researchers had some general ideas of what such
modes should be, they were not known with enough
precision for effective control.
Finally, the student came
up with procedures that could control such motions while
automatically accommodating processes such as docking and
undocking of capsules and the addition and consumption of
consumables, processes that can cause major changes in
the dynamic properties of space structures.
To try his system, Teague
built an ungainly-looking test bed that allows him to
simulate the movement of a light structure in
weightlessness. The test bed consists of three 100-pound
blocks of aluminum connected by two 15-foot long rods.
Each block has two arms that are about five feet long
extending perpendicular to the rods. On the end of each
arm is a small GPS receiver and a cluster of four
compressed-air thrusters.
The entire assembly is
hung by extremely strong thread. The top of each aluminum
block is milled out in a cone shape so that the thread
can be attached at its center of mass and the block can
rock without contacting the thread. Threads from each of
the three blocks extend upward where they are attached to
a 30-foot length of heavy steel pipe. Straps from each
end of the pipe are tied onto a thrust bearing that
allows the entire assembly to rotate. The bearing, in
turn, is supported by a heavy, overhead crane.
Because the rods
connecting the three blocks are extremely flexible, the
testbed can simulate a wide variety of motions. Each of
the blocks can be set rocking vertically and
horizontally. The rods transmit some of this motion to
the other blocks. So waves of motion can travel from one
end of the assembly to the other. When the blocks are set
rocking in different directions, the waves can combine
and cancel in unexpected ways.
Teague's test area is
indoors, so he had to use pseudo-satellites, antennas
that produce imitation GPS satellite signals. The
receivers use these signals to keep track of their
precise position. All the positions are sent to a desktop
computer that contains a model of the assembly. The
computer identifies the oscillation modes when they are
still very small and calculates the timing and duration
of the air blasts necessary to dampen them out.
The most dramatic
demonstration of the system's capabilities comes when
Teague vigorously sets the assembly rocking and rolling.
When he activates the control system, the thrusters begin
hissing, the motions get smaller and smaller and the
assembly returns to rest within 5 seconds.
A less dramatic, but more
realistic test cares when Teague turns the control system
on and then manually moves one of the arms. Thrusters
begin hissing immediately and the arm rapidly returns to
its proper position when he lets go.
Teague also can use the
system to rotate the flimsy assembly as if it were rigid.
When he enters the proper command, the thrusters begin to
hiss and the assembly begins to turn like a rigid body,
with very little shuddering or deviation from its base
configuration.
The research was funded by
the National Aeronautics and Space Administration. SR
|
