Design of adaptive Kalman consensus filters (a-KCF)

In this paper, we propose a continuous-time adaptive Kalman consensus filtering (a-KCF) scheme motivated by the work of Olfati-Saber, in which the author combines the Kalman filtering algorithm with a consensus mechanism to achieve distributed multi-agent systems to track and self-calibration in sensor networks. Using the adaptation law, we show the improvement of the convergence rate of the associated difference between the estimation of each agent and the actual states of a dynamic linear system. In other words, through this (a-KCF) protocol, we achieve the consensus among agents in a faster fashion than the non-adaptive (KCF) scheme. Simulation results of the performance of (a-KCF) confirm the validation of our design.