Distributed estimation design for LTI systems: a linear quadratic approach

Abstract

This paper deals with the problem of distributedly estimate the state of a plant through a network of interconnected agents. Each of these agents must perform a real-time monitoring of the plant state, counting on the measurements of local plant outputs and on the exchange of information with neighbouring agents. The paper introduces a distributed LQ-based design that is applied to a distributed observer structure based on a multi-hop subspace decomposition. Stability and optimality conditions are derived and tested in simulation. Finally, the design method presented allows the user, through the tune of two scalar parameters, to modify the observer gains according to their experience about the plant.