Average State Observers for Large-Scale Network Systems

In this paper, we propose a novel observer, called the average state observer, for large-scale network systems. This observer estimates the average behavior of the system with an estimation error assurance. To design an average state observer with explicit consideration of the estimation error, we first derive a tractable representation of the estimation error system. On the basis of this representation, we provide a theoretical upper bound of the estimation error for the average state observer. As a result, a systematic procedure to design the average state observers with an estimation error assurance is presented. Finally, we show the efficiency of the average state observer through an example of spatially discretized thermal diffusion networks.