Control-Theoretical and Topological Analysis of Covariance Intersection Based Distributed Kalman Filter

Resilient and consistent multirobot cooperative localization with covariance intersection

“…The second assumption is imposed to assure that the entire system configuration is stationary. As a result, the boundedness analysis of the cooperative localization algorithm can be achieved by that of the distributed estimation algorithm [35].…”

“…We now apply the result of the distributed Kalman filter with CI in [35] to analyze the covariance boundedness of Ψ i,t . To explicitly characterize the relations among all robots, we use graphs to describe the observation and the communication configurations in the multirobot system.…”

“…Definition 2 (Super Neighborhood [35]). For j = i, j ∈ S i if there exists a path in G C from j to i.…”

“…First of all, the equations ( 31) and ( 32) are exactly those equations ( 8) and ( 12) in [35]. Therefore, we can consider the upper bound of the cooperative localization Ψ t as a realization of the distributed estimation algorithm described in [35], whose boundedness criterion has been established.…”

Sequential covariance intersection‐based Kalman consensus filter with intermittent observations

“…Furthermore, a systematic way to prove the boundedness of the estimation error and the estimation error covariance for distributed Kalman filters in general time‐varying systems was presented. More recently, Chang and Mehta [25 ] investigated the relationship between effective observability in each agent and system topology, and discovered that some CI links may not provide an improvement on estimation performance but generate additional uncertainty. However, these consensus weights calculated by CI are based on the batch processing technique, which inevitably increases the computational burden and degenerate the performance for some practical applications.…”