Robust receding-horizon estimation for uncertain discrete-time linear systems via semidefinite programming

A novel approach for the solution of min-max receding-horizon state estimation problems for a class of uncertain linear systems is proposed that is based on semidefinite programming. Suboptimal solutions are sought for which a certain error is admitted with respect to the optimal cost value, thus ensuring the numerical tractability. For such approximate solutions, sufficient conditions are given for the stability of the estimation error dynamics in the presence of bounded disturbances and explicit bounding sequences are provided.