Robust recursive estimation for the errors-in-variables nonlinear systems with impulsive noise

The non-Gaussian characteristic of the external disturbance poses a great challenge for system modeling and identification. This paper develops a robust recursive estimation algorithm for the errors-in-variables nonlinear system with the impulsive noise. The algorithm is formulated by minimizing the continuous logarithmic mixed p-norm criterion, and is capable of giving a robust estimation against the impulsive noise through an adjustable weight gain. The nonlinear monomials of the noisy input are estimated by the recursive expressions based on the bias correction. Furthermore, a continuous logarithmic mixed p-norm based robust hierarchical estimation algorithm is derived to reduce the computational loads. The simulation studies demonstrate the feasibility of the proposed algorithms.