分组故障检测证据融合多星故障隔离算法

Translated title of the contribution: Multi-satellite faults isolation algorithm based on group fault detection and evidences fusion

Hui Bin Wang, Yong Mei Cheng, Wei Fan, Song Li, Bin Gou

Research output: Contribution to journalArticlepeer-review

1 Scopus citations

Abstract

In allusion to the increased possibility of multi-satellite faults resulting from the increase of navigation satellite constellation, a multi-satellite faults isolation algorithm based on group fault detection and evidences fusion is proposed. Firstly, all visible satellites are divided into groups and each group is taken as an evidence, and the fault and fault-free belief assignment of every evidence is calculated by fault detection. Then, a reliability based weight calculation method is proposed to realize the multi-evidence fusion in the weighted distribution method, so the situation that fusion result varies with fusion order can be deal with, and the fault and fault-free belief assignment of current satellites can be obtained. Finally, the fault satellites are isolated by taking the mean of all single satellite fault belief assignment as the fault isolation threshold. The relationship between faults isolation effect and group number of the proposed method is analyzed in GPS/BDS navigation mode. Furthermore, contrastive analysis is done among the faults isolation effect of the random search method, the quality value method and the proposed method, which verifies the effectiveness and superiority of the proposed algorithm.

Translated title of the contributionMulti-satellite faults isolation algorithm based on group fault detection and evidences fusion
Original languageChinese (Traditional)
Pages (from-to)1265-1270
Number of pages6
JournalKongzhi yu Juece/Control and Decision
Volume34
Issue number6
DOIs
StatePublished - 1 Jun 2019

Fingerprint

Dive into the research topics of 'Multi-satellite faults isolation algorithm based on group fault detection and evidences fusion'. Together they form a unique fingerprint.

Cite this