Dynamically Self-Configuring Automotive Systems (DySCAS) is a research project funded by the European Commission within Sixth Framework Programme (FP6) under Information Society Technologies (IST) priority. The project started June 1 2006 and will end in December 2008.
DySCAS targets an automotive system that emphasizes the dynamic self-configuration of embedded systems. The main goal of the project is the development of an intelligent automotive network middleware that, unlike common static design-time configurations, adapts itself to cater for an updated hardware or software component configuration, either off-line or online. This middleware will render a platform having properties that include a high degree of robustness and fault-tolerance due to its self adaptability at runtime.
The DySCAS vision can be described as follows
1. Vehicle Systems adapt themselves to changing environmental conditions
2. Optimal usage of system resources (memory, CPU, communication) and load distribution. For example, runtime migration of functions among control units and execution domains.
3. The vehicle uses additional resources in its (communication) range.
4. Secure, reliable and efficient inter-domain communication and collaboration
DySCAS Use Cases
DySCAS is exemplified by following four generic use cases
1. New device attached to the car. Eg. Automatic playing of music on vehicle audio system from user mobile device.
2. Integrating new software functionality. Eg. Automatic download and installation of new software.
3. Closed Reconfiguration. Eg. Transfer of software across the network in case of failure.
4. Resource Optimization. Eg. Reconfiguration for better usage of system resources.
The main objective of the DySCAS project is the specification of a middleware suitable for dynamic reconfiguration in the context of embedded vehicle electronic systems. To achieve this, the DySCAS project will develop or adapt the technologies required. The work will be divided into the following areas:
The DySCAS project will develop or adapt basic mechanisms and concepts for the dynamic reconfiguration of systems, ranging from algorithms for run-time configuration, middleware solutions, to basic software implementation technologies, e.g. operating systems and communication networks.
To integrate the aspect of dynamic self-(re)configuration into a system several architectural aspects must be considered. DySCAS will provide architectural patterns for a layered software architecture suitable for deployment in automotive control systems that supports the generic provision of self-management behaviours (which include self-configuration, self-healing, self optimisation, and self protection).
Methods and Tools
In the DySCAS project appropriate tool support and modelling techniques shall be developed supporting the development and evaluation of dynamically reconfigurable systems.
The DySCAS project will draw benefits both from established techniques in the industry and from state of the art research results. To ensure wide usability, the results will be developed in conformance with AUTOSAR.
The main outcome of DySCAS will be an open specification of an automotive middleware for dynamic reconfiguration. This specification will include findings regarding e.g. limitations in technology, cost versus functionality trade-offs
and performance estimates, that have affected the design of the middleware. This knowledge will be obtained through extensive simulations as well as a reference implementation of the DySCAS platform. Apart from the technical
evaluation, an important part will be the investigation of the DySCAS concept from a business point of view. A number of demonstrators will be built to
further validate the DySCAS platform and to showcase the functionality of the DySCAS middleware. Multidisciplinary integration of technologies such as, Policy Based Computing, Control Systems and Model Based Development is one of the important features of DySCAS.
Role of project partners
The consortium members represent a well chosen mix of OEMs, suppliers, research institutes and SMEs in order to ensure project success.
Volvo Technology is the project coordinator and will together with Daimler provide use cases and requirements for the services to be implemented in the middleware. As OEMs, these partners will also contribute with experience on electronic architectures for vehicles.
As suppliers, Enea and Bosch will provide knowledge and experience on middleware, operating systems, communication and resource management from both the automotive and telecommunications domain.
The universities will contribute with state-of-the art expertise required to deploy and develop the services to be supported by the DySCAS middleware. This include policy-based computing (The University of Greenwich), quality-of-service and real-time aspects (University of Paderborn), and architecture design and control-theoretic approaches to
computer control (Royal Institute of Technology).
The SMEs Movimento and Systemite will primarily contribute by building demonstrators for the DySCAS platform. Both companies have a background as tool suppliers for embedded systems.
Dr. Martin Sanfridson
Volvo Technology AB
Sven Hultins Gata 9C
SE-412 88 Göteborg, Sweden
Tel. +46 (0) 31 322 06 65
Fax. +46 (0) 31 772 40 86
KTH Team Leader
Prof. Martin Törngren
Tel: +46 - (0)8 7906307