Dynamic State Estimation in the Presence of Sensor Outliers Using MAP-Based EKF

In this letter, we consider the problem of dynamic state estimation (DSE) in scenarios where sensor measurements are corrupted with outliers. For such situations, we propose a filter that utilizes maximum a posteriori estimation in the extended Kalman filtering framework to identify and discard the outlier-ridden measurements from a faulty sensor at any given time instant. However, during this process, all those measurements that are not affected by outliers are still utilized for state estimation. Using an illustrative example of dynamic target tracking, we demonstrate the effectiveness of the proposed estimator.