A Game-Theoretic Approach to Robust Fusion and Kalman Filtering Under Unknown Correlations

摘要

This work addresses the problem of fusing two random vectors with unknowncross-correlations. We present a formulation and a numerical method forcomputing the optimal estimate in the minimax sense. We extend our formulationto linear measurement models that depend on two random vectors with unknowncross-correlations. As an application we consider the problem of decentralizedstate estimation for a group of agents. The proposed estimator takescross-correlations into account while being less conservative than the widelyused Covariance Intersection. We demonstrate the superiority of the proposedmethod compared to Covariance Intersection with numerical examples andsimulations within the specific application of decentralized state estimationusing relative position measurements.