AI Planning and Scheduling for Autonomy in Space Application
------------------------------------------------------------
Saturday July 6th, Manchester
http://www.nicve.salford.ac.uk/space/
In association with TIME 2002: 9th International Symposium on
Temporal Representation and Reasoning
Sponsored by Science Systems
Background
----------
Spacecraft and satellites have typically been supported by large numbers of
operations staff on the ground, who controlled every aspect of their
activity by generating very specific and lengthy sets of detailed low-level
commands. With the possibility of losing a mission if an error was made,
this was seen as the safe way of operating.
However a number of factors make it hard to continue like this. One is
cost: a comprehensive network of ground stations operated round the clock
is not cheap. This combines with a scaling up problem: as more satellites
and spacecraft are launched and commercial operations increase, multiplying
the cost to an unacceptable level. 'Lights out' operation, where ground
stations are operated only during office hours is already coming in. A
second factor is more ambitious missions: constellations of satellites,
more deep space probes where communication delays are a factor, and the
long-term aim to put people on Mars. The third factor is the desire to have
scientists interact with their space-based experiments directly, perhaps by
extending the internet into space.
For all these reasons, autonomous operation of satellites and spacecraft is
becoming more interesting to the Space community. AI Planning and
Scheduling has an important role to play in providing on-board autonomy,
but for it to succeed, researchers need to understand the requirements of
Space applications and Space people need to understand what the technology
can achieve. This workshop hopes to develop the dialogue between the two
communities.
A proposal for a journal special issue based on submissions will be
formulated.
Submissions
-----------
We request submissions both from researchers in AI Planning and Scheduling
who are interested in applying their ideas to the Space domain, and from
companies and people working on Space applications discussing possible
applications areas where intelligent planning and scheduling would be
useful.
Relevant topics include:
o Continuous planning and replanning; Dynamic scheduling
o Planning and scheduling under real-time constraints
o Efficient temporal planning
Efficient planning under uncertainty
o Integration of planning, scheduling and fault detection and diagnosis
Mixed-initiative planning and adjustable autonomy architectures
Distributed multi-agent planning
o Automating earth observation requests on satellites
o Autonomous operation and control of satellite constellations
Autonomous operation and control of planetary rover fleets
o Autonomous operation and control of deep space spacecraft
Verification of autonomous control systems
Any discussion of general techniques should be linked to how they might be
applied in the Space domain.
Please submit a maximum of 5000 words by the deadline below to:
Ruth Aylett Email: [log in to unmask]
CVE, Business House
University of Salford
Salford M5 4WT
Electronic submission heavily preferred: pdf (most preferred), MSWord, or
postscript
Important dates
---------------
Submissions Friday May 10th
Notifications Monday Jun 3rd
Deadline for camera-ready copy Monday Jun 17th
Programme committee
--------------------
Ruth Aylett University of Salford (Chair)
Eric Bornschlegl, European Space Agency, ESTEC
Amadeo Cesta IRNS Rome
Mike Fisher University of Liverpool
Maria Fox Durham University
Tim Grant Origen Software
Lina Khatib NASA Ames Research Center
Derek Long Durham University
Robert Morris NASA Ames Research Center
Mark Woods Science Systems
Ruth Aylett Professor of Intelligent Virtual Environments
email: [log in to unmask] phone: 44-161-295-2912
Home page: http://www.nicve.salford.ac.uk/ruth/
fax: 44-161-295-2925
Centre for Virtual Environments, Business House,
University of Salford, Salford, M5 4WT, UK 'Life is beautiful.'
