Optimal and Steady-State Multisensor Covariance Intersection Fusion Kalman Filters

Abstract

Listen

Translate article

This chapter presents the optimal and steady-state covariance intersection (CI) fusion Kalman filters with unknown cross-covariances among the local filtering errors for the multisensor linear discrete time-invariant stochastic system,. Their accuracies are higher than those of the corresponding local Kalman filters, and lower than those of the corresponding optimal and steady-state fused Kalman filters with known cross-covariances. Using the existence theorem of implicit functions and the continuity of a function, it has been rigorously proven that the steady-state CI-fused Kalman filter converges to the optimal CI-fused Kalman filter in a realization. One simulation example for a tracking system with a three-sensor verifies its accuracy relations and shows the convergence.