A Square Root Extended Set Membership Algorithm with Applications to Nonlinear System Estimation

In this paper, a square root extended set membership (SRESM) state estimation algorithm is proposed to further improve the numerical accuracy and stability of the nonlinear system estimation. In comparison with the existing extended set membership (ESM), the proposed square root ESM using the efficient and stable updating recursions provides much more accurate estimation results. It has improved numerical stability and real time applicability. Simulation results in estimation of mobile robot localization are given to show the effectiveness and robustness of the proposed algorithm.