Measurement Outlier-resistant Mobile Robot Localization

This paper is concerned with the measurement outlier (MO)-resistant mobile robot localization (MRL) problem. For the purpose of mitigating the effect of the MOs, a time-varying state estimator is constructed containing a saturation function with variable saturation level. The purpose of this paper is mainly to seek an effective solution to the addressed MRL problem by devising the desired time-varying state estimator which ensures that, over a finite horizon, the estimation error dynamics satisfies the H ∞ performance constraint. By constructing an appropriate Lyapunov function, the existing condition of the estimator is first obtained. Then, the desired state estimator gain is given through the solution to a set of certain matrix inequalities and the MO-resistant MRL algorithm is presented. Finally, an example is conducted to testify the usefulness of the MRL algorithm proposed.