With the advancement of digitalization, AI, and modern data communication, engineering faces growing challenges from complex systems-of-systems, where perception sensors process data independently, information is merged centrally, and cross-platform data exchange is managed holistically. As a result, distributed architectures for data fusion, state estimation, and multi-target tracking have become increasingly important. Using multiple sensors to build situational awareness generates large volumes of data that must be filtered, enriched, interpreted, and evaluated—tasks that are only feasible with distributed algorithms.
Theory and Methods for Distributed Data Fusion Applications begins by covering the fundamentals of stochastic motion, which underpins models for target tracking and parameter estimation. It then introduces key target tracking and state estimation algorithms, followed by methods for estimating full trajectories using data from multiple time points. The book next addresses the problem of distributed fusion and presents initial solutions. The final chapters discuss track fusion with unknown cross-covariances, the Distributed Kalman Filter and its variants, and methods for track-to-track association.
This book offers engineers state-of-the-art approaches and algorithms for data fusion in distributed systems, providing a deep understanding of the constraints, assumptions, advantages, and limitations. Each method is derived in detail, with all auxiliary calculations and intermediate steps included.
