| dc.description.abstract | Artificial Intelligence is one of the most demanding new research areas and just “the first
small step” has been taken whilst the “giant leap for mankind” is still outstanding.
In the meantime enormous efforts have been undertaken to try to find out in which area and
how far supporting and self-learning elements could be defined and should be introduced in
order to finally copy the status or imitate -more or less- a human being, its behaviour,
reactions, feelings and especially its ability to continuously broaden one’s own horizon of
knowledge independently of the actual environment and situation.
In this context space and the human research curiosity about deep space exploration
resulting in plans for long duration human space missions indispensable requires adequate
intelligent supporting tools.
These tools may be simple MMI instruction, simple expandable technical tool boxes in sense
of a technical data base, personal electronic partners, complex ontology data bases with
medical and behavioural data and of all kinds more.
Space operations are of high complexity due to the numerous interfaces and difficult to
manage, even in the nominal cases, let alone in off-nominal situations (see current situation
on the next page and cover page image). We can expect that for future human (deep) space
missions this complexity will only increase. And the greater the distance between Earth and
the spacecraft will be, simply due to the technical effect of the runtime delay of the radio
signal (about 141 minutes in one direction to MARS) effective assistance in all different areas
of human aspects will be of relevance.
It thus will be crucial to have the support of a Knowledge Interoperability Infrastructure (KII)
that will be able to provide the right information at the right time to the right person. The
context for this is a multi-user environment, where it is expected that geographically (and
perhaps even temporally) distributed teams, consisting of various combinations of humans
(crew, flight control, scientific experts, etc.) and software (reasoning agents, smart systems
and instruments, robots, crew ePartners, etc.) work together on common goals and
objectives.
The focus my thesis, the MECA-lite project, which is based on the KII (section 3.2) and
MECA (section 3.1) project, therefore is on investigating the considerations and
requirements to identify relevant use cases and scenarios and to propose a suitable
formalism for Knowledge Interoperability and Representation. This formalism is then
evaluated in practice by building an example knowledge base that uses it, and showing how
the information present in this knowledge base can be used to add value to the humanmachine
teams working on solving common problems. | |
