Intelligent Sensor Fusion for Shipboard Auxiliary Systems

Data analysis and sensor fusion is undeniably the most pertinent part of science and practical applications related to information management for damage detection and condition monitoring. Unfortunately, for a general solution to be tractable for coupled, distributed systems, such as shipboard auxiliary systems, sensor fusion requires innovative techniques and algorithms. The ultimate goal for the proposed sensor fusion is to address the problem of damage detection in the auxiliary systems, to improve situational awareness, and to formulate appropriate control actions. Our technique combines a statistical signal processing approach based on Hidden Markov Modeling with nonlinear estimation theory developed for complex distributed systems. Fusion of data from multiple sources will lead to managing the information regarding sensor features simultaneously. We aim to achieve a reliable and computationally inexpensive sensor fusion technology targeted for shipboard auxiliary systems.