Introduction - Autonomous and Autonomic Systems - page 12

Table 1.2. Various similar three-tier approaches in different domains

DARPA/ISO's
autonomic
information
assurance

Self-directing and
self-managing
system
potential

Intelligent
machine
design

Future
comms.
paradigm

NASA's
science
mission

Reflection Knowledge

plane

Mission
plane

Science Autonomous

Routine Management

control
plane

Cyber
plane

Mission Selfware

Reaction Data

plane

Hardware
plane

Command
sequence

Autonomic

mechanisms for information systems against malicious adversaries. The pro-
gram developed an architecture consisting of three planes: mission, cyber, and
hardware. One finding from the research was that fast responses are necessary
to counter advanced cyber-adversaries [ 87 ], similar to a reflex action discussed
earlier.
As will be defined later in this topic, NASA's science mission management,
from a high level perspective, may be classified into:
Science planning: Setting the science priorities and goals for the mission.
Mission planning: Involving the conversion of science objectives to instrument
operations and craft maneuvering, housekeeping and scheduling, comm-
unications link management, etc.
Sequence planning: Production of detailed command sequence plans.
These versions of a high-level, three-tier view of self-governing and self-
managing systems may be generalized into autonomous-selfware-autonomic
tiers. Of course, this is intended neither to be prescriptive nor to be in con-
flict with other views of autonomic systems. The intention in examining and
viewing systems in this light is to assist in developing effective systems.

1.3 Using Autonomy to Reduce the Cost of Missions

Spacecraft operations costs have increasingly concerned NASA and others and
have motivated a serious look into reducing manual control by automating as
many spacecraft functions as possible. Under current designs and methods
for mission operations, spacecraft send their data (engineering and science)
to earth for processing and receive commands from analysts at the control
center. As the complexity and number of spacecraft increase, it takes a pro-
portionately larger number of personnel to control the spacecraft. Table 1.3
shows some current and future missions with the number of people needed
to operate them [ 122 ]. People-to-spacecraft ratios are shown (a) for past and
present missions based on current technology, and (b) for expected future mul-
tispacecraft missions with the current technology and operations approaches.