Optimally Robust Kalman Filtering at Work: AO-, IO-, and Simultaneously IO- and AO- Robust Filters

TL;DR

This paper develops a hybrid robust Kalman filter that simultaneously handles additive and impulsive outliers, improving state estimation robustness in systems with mixed outlier types.

Contribution

It introduces a novel hybrid filter combining AO- and IO-robust procedures to effectively manage both outlier types at once.

Findings

Hybrid filter outperforms traditional methods in mixed outlier scenarios

Demonstrates robustness in a reference state space model

Compared favorably with ACM and median-based filters

Abstract

We take up optimality results for robust Kalman filtering from Ruckdeschel[2001,2010] where robustness is understood in a distributional sense, i.e.; we enlarge the distribution assumptions made in the ideal model by suitable neighborhoods, allowing for outliers which in our context may be system-endogenous/propagating or -exogenous/non-propagating, inducing the somewhat conflicting goals of tracking and attenuation. Correspondingly, the cited references provide optimally-robust procedures to deal with each type of outliers separately, but in case of IO-robustness does not say much about the implementation. We discuss this in more detail in this paper. Most importantly, we define a hybrid filter combining AO- and IO-optimal ones, which is able to treat both types of outliers simultaneously, albeit with a certain delay. We check our filters at a reference state space model, and compare the results with those obtained by the ACM filter Martin and Masreliez[1977], Martin[1979] and non-parametric, repeated-median based filters Fried et al.[2006,2007].