State estimation strategies are vital for obtaining knowledge of a dynamic system’s state when faced with limited measurement capability, sensor noise, or uncertain system dynamics. The Kalman filter (KF) is one of the most widely recognized filters and provides the optimal solution for linear state estimation problems. The smooth variable structure filter (SVSF) is a model-based strategy which is also formulated as a predictor-corrector. Despite being a suboptimal estimator, the SVSF is highly robust to modeling uncertainties, errors, and system change. The combination of the SVSF with the KF (SVSF-KF) results in an adaptive estimation algorithm which provides an optimal KF estimate in normal operating conditions, and a robust SVSF estimate in the presence of faults or uncertainties. While effective in some cases, the SVSF-KF has been shown to suffer from several drawbacks associated with the time-varying smoothing boundary layer and adaptive gain used to detect system change. Several new approaches have been proposed in recent years with the aim of improving the SVSF-KF’s performance. Among these approaches is a novel gain formulation based on the normalized innovation squares, while another makes use of the interacting multiple model framework. In this paper, we review the newly proposed SVSF-KF formulations and compare their performance on an electro-hydrostatic actuator test case.