Distributed Kalman filter for linear system with complex multi‐channel stochastic uncertain parameter and decoupled local filters

Abstract

Summary This article considers a distributed Kalman filtering problem for linear system contaminated by complex multi‐channel random uncertain parameter in which a number of nodes cooperative without central coordination to estimate a common state based on local measurements and data received from neighbors. We propose an approach to eliminate this error propagation. The proposed local filters are guaranteed to be stable under some mild conditions on certain global structural data, and their fusion yields the centralized Kalman estimate. Then, we extend this method to smoothing and deconvolution algorithm. Finally, simulation experiments demonstrate the validity of the proposed approach.

Consensus‐based robust least‐squares filter for multi‐sensor systems

“…where Q is defined the same as (17), and Ŵ is in the form of a block diagonal matrix (18) with the same ŵ22 , and the following ŵ11 ,…”

“…A consensus algorithm determines the exchange of information between all neighbors in a network and results in reaching an agreement on a certain quantity or common value in all sensors 14 . Recently, consensus control and the integration of multiple sensors and consensus filters has been reported in considerable amount of publications 15‐18 . The theory, design method, and problem‐solving of consensus filtering in a network of systems are presented in References 19 and 20.…”

“…14 Recently, consensus control and the integration of multiple sensors and consensus filters has been reported in considerable amount of publications. [15][16][17][18] The theory, design method, and problem-solving of consensus filtering in a network of systems are presented in References 19 and 20. In general, based on the agreement parameter, consensus filtering approaches are classified into four subgroups.…”

Experimental Validation: Perception and Localization Systems for Autonomous Vehicles using the Extended Kalman Filter Algorithm

“…The KFs technique operates under the assumption of Gaussian noise impacting data, which might not always hold universally valid in this context. Despite their impressive convergence attributes, KFs are initially designed to address linear system challenges and are seldom applied in SLAM [16,17]. However, the practical problems of autonomous vehicles in the real environment tend to be nonlinear systems due to the complexity of the environment and the uncertainties [18].…”

Trajectory Modeling by Distributed Gaussian Processes in Multiagent Systems

“…Generally speaking, the processing is called distributed manner if it is carried out by a cooperative strategy among nodes without central coordination [ 19 ]. The distributed method aims at minimizing the amount of computation and communication required by each node as well as making these requirements scalable in the number of nodes [ 20 ]. Distributed methods are available for parameter estimation [ 14 ], Kalman filtering [ 21 ], control [ 22 ], optimization [ 23 ], learning [ 13 ], etc.…”