Distributed filtering basing consensus for the local strongly coupled systems

The interactions between subsystems are important for large-scale systems. We introduce a local strongly coupled system which coupled by random communication between subsystems.Due to the intermittent communication, it is difficult to apply the standard Kalman or robust filter to design procedures to such systems. In this paper, we addressed the distributed robust filter design method for this kind of system based on the consensus idea. The main result is a sufficient condition which guarantees a suboptimal level of disagreement of estimates in a coupled network of estimators. The condition is formulated in terms of feasibility of biaffine matrix inequalities (BMIs). The generic algorithm is used to treat the bilinear relation between filter parameters and the interconnection gains. The proposed approach is applied to the problem of formation-based robust synchronization. The numerical simulations show the effectiveness of the proposed filtering method.