Environmental Engineering Reference
In-Depth Information
6.
Data monitoring and trending (as discussed)
7.
“Smart” fault detection, diagnosis, isolation, and correction (as discussed)
8.
Look-ahead modeling (a function of planning and scheduling autonomy)
9.
Target planning and scheduling (as discussed)
10.
SI commanding and configuration (execution of science and calibration
requests)
11.
SI data storage and communications (in cooperation with ground agent)
12.
SI data processing (as discussed)
6.6.6 Deep Space Missions
Deep space missions are tailor-made for the flexibility and responsiveness en-
abled by a Remote Agent implementation of an autonomous spacecraft. Being
out of contact with the ground for very long periods and with significant radio-
signal propagation time for communications, the spacecraft needs to have a
greater degree of “self-awareness” not only to maintain H&S, but even to
perform its mission eciently. This is particularly true of a mission like an
asteroid flyby where mission-critical observing decisions must be made in re-
altime. Although very complex deep space missions have successfully been
performed by Jet Propulsion Laboratory (JPL) in the past, JPL has rather
clearly determined that the key to maintaining their recent downward trend
in mission cost is to promote steadily increasing onboard autonomy through
the use of formalisms such as Remote Agents. In the DS1 mission, the re-
sponsibility of health monitoring was transferred from ground control to the
spacecraft [
99
,
135
,
195
]. This marked a paradigm shift for NASA from its
traditional routine telemetry downlink and ground analysis to onboard health
determination.
6.6.7 Spacecraft Constellations
Being recent in development, where deployment for now is largely restricted
to communications networks, spacecraft constellations represent a new oppor-
tunity for flight autonomy applications and are covered in Chap.
9
. Whereas
consideration of Remote Agent applications to the previous mission types
concerned interactions of spacecraft subsystems with each other or with the
ground, for constellations, the scale of interaction expands to conversations
potentially between all the various members of the constellation. In a complex
“conversation” of this type, just the job of determining which members of the
constellation should be included in the conversation, when they should enter
the interaction, and when they should drop off can be a thorny problem. More
discussion on this topic will be provided later, but to support constellation in-
teractions, a hierarchy of subsystem subagents controlled by agent spacecraft
will need to be introduced.
