Improved Gaussian filtering for handling concurrent delayed and missing measurements

Abstract

This article addresses the Gaussian filtering problem under the environment of jointly occurring delayed and missing measurements. In this work, the former irregularity is incorporated (in the measurement model) using a Bernoulli random variable (BRV) and a geometric random variable, while the latter is subsumed with the help of the BRV; thereby, it enables to take account of large delay extents efficiently. Specifically, a modified measurement model, which incorporates the concerned irregularities, is introduced. Accordingly, the measurement‐related statistical parameters, that is, measurement estimate, covariance, and cross‐covariance, are rederived with respect to the modified measurement model. The rederived parameters replace the corresponding ones in the traditional Gaussian filtering algorithm, resulting in the proposed Gaussian filter. The simulation results conclude the superior performance of the proposed filter over the existing filters in handling the coexisting delay and missing measurement irregularities.