Abstract:

To solve the constraints of packet dropouts, quantization and model uncertainty in networked systems for state estimation, a robust moving horizon estimation (MHE) algorithm with prediction compensation is proposed. A group of Bernoulli distributed random variables is employed to describe the phenomenon of packet dropouts and the predictor of the missing measurements is applied as a compensator, the error introduced by data quantization is described as a bounded uncertainty parameter in the observation equation, the uncertainty of the model is described by stochastic parameter perturbations in the system matrix, based on the moving horizon strategy, and considering the worst-case caused by quantization and model uncertainty, the optimal state estimation is obtained by solving a min-max problem. The stability of the proposed algorithm is studied, explicit bounding sequence on the expectation of the square norm of estimation error is obtained, and a sufficient condition for the convergence of the square norm of estimation error is given. Finally, an example is given to demonstrate the efficiency of the proposed method.

Key words: moving horizon estimation (MHE), prediction compensation, quantization, model uncertainty, min-max problem, stability analysis