Multisensor Parallel Largest Ellipsoid Distributed Data Fusion with Unknown Cross-Covariances

Abstract

As the largest ellipsoid (LE) data fusion algorithm can only be applied to two-sensor system, in this contribution, parallel fusion structure is proposed to introduce the LE algorithm into a multisensor system with unknown cross-covariances, and three parallel fusion structures based on different estimate pairing methods are presented and analyzed. In order to assess the influence of fusion structure on fusion performance, two fusion performance assessment parameters are defined as Fusion Distance and Fusion Index. Moreover, the formula for calculating the upper bounds of actual fused error covariances of the presented multisensor LE fusers is also provided. Demonstrated with simulation examples, the Fusion Index indicates fuser’s actual fused accuracy and its sensitivity to the sensor orders, as well as its robustness to the accuracy of newly added sensors. Compared to the LE fuser with sequential structure, the LE fusers with proposed parallel structures not only significantly improve their properties in these aspects, but also embrace better performances in consistency and computation efficiency. The presented multisensor LE fusers generally have better accuracies than covariance intersection (CI) fusion algorithm and are consistent when the local estimates are weakly correlated.

An Omnidirectional Morphological Method for Aerial Point Target Detection Based on Infrared Dual-Band Model

“…Since the probability of detection of the dual-band model is obviously superior to that of the single-band model according to the simulation result in Section 2.3, all the compared algorithms are performed based on the proposed dual-band model. The probability of detection, the probability of false alarm and the running time of the target detection algorithm are selected as the evaluation indexes of the results, which are computed from the following: (25) where N c is the number of detected true point targets, N f is the number of false alarms, and N t is the total number of point targets.…”

“…In recent years, multisensor data fusion is widely applied in airborne sensing, medical diagnosis, and disaster prediction [23][24][25][26]. Mehmood et al [27] presented the wavelet and Reed-Xiaoli (RX) algorithm for dual-band forward-looking infrared imagery, but it did not work for very small-sized targets.…”

Hierarchical Decentralized State Estimation with Unknown Correlation for Multiple and Partially Overlapping State Vectors

“…Therefore, similar to EI, it does not guarantee consistency [28] unless the local estimates are weakly correlated [24]. However, instead of using a sequential approach, the aim is to compute static weight matrices W i such that the global state estimate becomes,x…”

“…The weights W i in (24), and (25) will be obtained by solving two semidefinite programming (SDP) problems. Proposition 5.1: [29] An inner ellipsoidal approximation of the intersection for the ellipsoids H 1 Y 1 H T 1 , ..., H n Y n H T n , can be obtained by solving the following Linear Matrix Inequality (LMI):…”

“…The framework of the three fusion methods (CI, EI, and LEA) have been devised for fusing two estimates only. When extended to multiple estimates the general recommendation is to sequentially apply these fusion methods [23], [24]. However, as demonstrated in [25] when using CI for three or more estimates it tends to overestimate the covariance, and therefore, the minimum overestimate no longer holds.…”