Environmental Engineering Reference
In-Depth Information
9.4.1 Ground-Based Constellation Autonomy
Figure
4.7
(Sect.
4.3.4
,p.
89
) shows a high-level representation of a constella-
tion simulation of ground-based autonomy for a constellation of four satellites.
In this simulation, a number of agents are connected to an environment in
which the ground control systems and satellites [
177
,
183
] are simulated. In
the simulation, the satellites are in orbit collecting magnetosphere data. The
simulation environment propagates the orbits based on ideal conditions. Faults
can also be inserted into the telemetry stream to simulate an anomaly.
The group of surrogate spacecraft agents, as a major component of the
ground-based community, maintains an awareness of the actual physical con-
stellation. The surrogates act on behalf of their respective spacecraft in status
monitoring, fault detection and correction, distributed planning and schedul-
ing, and spacecraft cooperative behaviors (as needed).
A next phase in the evolution of the above ACT scenario will be to have
communities of agents, each associated with a particular spacecraft in the con-
stellation. Each of these communities would have specialist subsystem agents
that would monitor the various subsystems of the spacecraft and cooperate
with one another in the handling of anomalous situations. An overall coor-
dinator, or spacecraft agent, would lead the community and represent the
spacecraft to ground controllers. It would also represent the spacecraft to
other spacecraft agents in the constellation community for activities such as
distributed planning and scheduling, and other forms of collaboration.
In the context of spacecraft constellations, the ground-based group of sur-
rogate agents illustrates two major themes in our discussion: (1) surrogate
agents can indeed support the concept of constellation autonomy in a mean-
ingful way, and (2) having a ground-based community of surrogate agents
allows developers and users (controllers) to gain confidence and trust in the
approach.
9.4.2 Space-Based Autonomy for Constellations
Constellation autonomy (as opposed to a single spacecraft autonomy) corre-
sponds to an intrinsic property of the group. Constellation autonomy would
not apply to a fleet of spacecraft where each is operated without reference
to the others (e.g., a tracking and data relay satellites (TDRS) servicing the
communications needs of many user spacecraft). However, if each satellite
manages itself in the common external environment so as to maintain a group
functionality relative to the environment, even when the members of the group
do not communicate directly with one another, and the group accomplishes
the end purpose of the constellation, then the constellation can be considered
as a self-managing one, i.e., autonomous. This is considered justifiable, from
the mission perspective, because the group accomplishes the end result as a
system even though its members do not intentionally interchange information.
We make this distinction because of the question of autonomy and viability.
