| |
 |
Faster and more economic
computing devices, sophisticated sensor technologies and advanced
communication including wireless networks form today's technological basis
to think about cooperative systems which support large real-time
applications. These systems will comprise myriads of components,
themselves with significant embedded functionality. Key characteristics of
these applications include awareness of the physical environment and
capability of interacting with it, geographical dispersion, time and
safety criticality, mobility and the ability for long term system
evolution. Autonomy of components and decentralized control are
indispensable to put these systems to work.
The
focus of the workshop is: "how to
manage uncertainty and complexity of modular, distributed and potentially
large-scale systems, which otherwise must possess timeliness properties,
including forms of hard real-time behavior".
Challenges
to be addressed by the workshop:
 | Dependable models of temporal adaptation |
For the envisaged class of
systems it may be impossible to define precisely the operational
environment. Because of the need to cooperate dynamically over
inherently unreliable links, temporal uncertainty of cooperation has to
be considered. We may not be able to handle uncertainty by worst case
assumptions. On the other hand, environment interaction requirements may
dictate the need for some form of real-time hardness. In consequence,
new dependably adaptive models are sought to achieve temporally
consistent interactions.
Topics:
flexible temporal guarantees for interactions; models of temporal
consistency.
| System construction from autonomous
components |
Autonomy of components is
recognized as a means to construct robust systems and handle system
evolution. It is also seen as the basis to hide low level temporal
issues and compose a system of well defined entities in the functional
and temporal domain. This view is well supported on the technological
side by the emerging use of smart sensors and actuators which
incorporate computational and communication capabilities. In the design
process for building large scale systems, component-based system design
is a research topic and already has reached practical exploitation.
Concerning operational
aspects at the distributed system level there still is a need to address
appropriate communication models and component interfaces to encapsulate
temporal, dependability and functional properties
Topics:
architectures for interacting among autonomous components; component
interfaces; self-description of components and dynamic interactions.
| Control of information dissemination |
To support dynamic and
spontaneous interactions of autonomous components, active dissemination
and notification schemes have been recognized as appropriate
communication models. New real-time communication paradigms are emerging
which allow to confine propagation of information by thematic, temporal,
geographical, security, or physical scopes. This allows to control the
trade-off between global accessibility of information and the quality
required for a local interaction of components.
Topics:
encapsulating and scoping; filters and aggregators.
| Embedded systems middleware |
Particularly, the
abstractions and services of advanced middleware technology play a
decisive role to manage the complexity of embedded systems. The attributes
form the makings of a new breed of such systems, or better put,
embedded systems-of-systems, based on visions like pervasive and ad-hoc
computing, or ambient intelligence.
Topics:
time- and event-based middleware; sentient object platforms; protocols
for smart transducers; wireless infrastructures.
The workshop addresses research
on such visionary systems taking place in Europe, US and Asia/Pacific. The
goal
of the workshop is to:
| bring together researchers in the area |
| present a range of ongoing projects,
activities and results on these topics |
| stimulate the discussion about the
challenges in this area of research |
|
|
