The Kalman filtering for a class of local strongly coupled systems

This paper addresses a Kalman filtering problem for a class of local strongly coupled systems which are derived from complex systems with inter communication or constraint between nodes. To start with, a multi-agent system is introduced and the communication matrix is characterized by employing random series with Bernoulli distributions. Using augmentation techniques and stochastic methods, the Kalman filtering algorithm is deducted. The experimental results not only demonstrate the effectiveness of the proposed filtering method, but also show the inter-agent communication could improve the tracing effect in the multi-agent collaborative systems.