The aim of the project is to develop an intelligent data processing framework for analyzing and interpreting data collected from systems of systems comprising sensor and actuator networks such that faults are promptly detected, isolated, identified and accommodated for in future decisions or actuator actions. The aims of the iSense project will be achieved through the following scientific and technological objectives:

• To develop a rigorous formulation for cognitive fault-diagnosis and fault-tolerant control problems;
• To design cognitive fault-diagnosis schemes that can be effectively applied to monitoring and control applications of uncertain distributed environments;
• To develop a set of adaptation and learning algorithms that can be incorporated into the cognitive fault-diagnosis and fault-tolerant control schemes;
• To investigate the design, analysis and evaluation of fault-tolerant control schemes;
• To integrate the various components and build a system prototype for the iSense Platform for intelligent building applications.

The proposed framework provides a new paradigm for data processing in the presence of fault events for unstructured environments. Most approaches for real-time fault-diagnosis so far have focused on well-defined and structured environments, where relatively accurate mathematical models are available. The methods proposed here are based on adaptation and learning schemes, which will enhance the effectiveness and robustness of autonomous diagnostic systems by exploiting key correlations between measured variables, in both space and time. Furthermore, the proposed cognitive fault-diagnosis framework will facilitate more timely response to fault events either autonomously though the use of feedback control and fault accommodation techniques implemented via the actuation devices or through interaction with people.