Best Achievable Performance of Cooperative Kalman Filters for Decoupled Multi-agent Systems under Environmental Disturbances

Abstract This paper deals with a distributed state estimation problem for a class of multi-agent systems. The agents are assumed to be decoupled and identical, but affected by a common environmental disturbance in addition to individual process noises. In the presence of such a disturbance, the resulting optimal filter consists of decoupled steady-state Kalman filters with information exchange. The changes in the performance caused by the existence of the environmental noise and the required information exchange are evaluated based on block decomposition of the stabilizing solution of the associated algebraic Riccati equation. Moreover, an optimal filter for the system with an additional global measurement output is derived. The performance improvement due to the additional measurement is also evaluated.