Semidefinite programming solutions to robust state estimation problem with model uncertainties

In this paper, a novel finite-horizon, discrete-time, time-varying state estimation method is proposed based on the recent robust semidefinite programming (RSDP) technique. The proposed formulation guarantees a robust performance with respect to model uncertainties which are known to lie within certain a priori bounds. This is in contrast to earlier robust designs, such as H^∞, which accommodate all conceivable uncertainties and therefore lead to overly conservative solutions.