An iterative partition-based moving horizon estimator with coupled inequality constraints

We propose an iterative, partition-based moving horizon state estimator for large-scale linear systems that consist of interacting subsystems. Every subsystem estimates its own state and disturbance variables, taking into account the estimates received from neighboring subsystems. Compared to other partition-based moving horizon estimators, the proposed method has two unique features: it can handle coupled inequality constraints on the estimated variables and its state estimates come arbitrarily close to the optimal state estimates of a centralized moving horizon estimator. The applicability and performance of the proposed method are demonstrated on a numerical example and convergence and asymptotic stability are rigorously proven.