AIS: Autonomous Integrated Systems

The characteristics of electronic systems to react autonomously and flexibly to faults and environmental modifications require a new kind of thinking in the design process. Sensors, evaluators and actuators in MPSoCs are necessary to detect and analyze spontaneous faults and to initiate actions that guarantee a reli-able operation beyond faults and interferences. Self-calibrating techniques and encoding procedures are promising approaches to increase reliability.

The goal of AIS is to research a new kind of design methodology for autonomous integrated systems. With this new design methodology components of architectures will be dimensioned with autonomous characteristics and provided for system design. In a process of exploration and integration of these components a combined autonomous design system methodology will be proposed. The design process will include – besides autonomous components and communication – an approach to consider also the operating system.

As a result based on such new design methodology, MPSoCs autonomously will find errors, faults and interferences to enhance the reliability of single components and complete systems. To do this, autonomous elements will be included to monitor data-paths, control-paths and structure of communication on functional level and to react to changed operating conditions and faults. The robustness will be optimized by inserting self-optimizing services including an operating system that is able to replicate, to monitor and to migrate tasks and services

Classification in the edaMatrix:

Used Abbreviations