Real-time fault detection method based on belief rule base for aircraft navigation system

• DOI: 10.1016/j.cja.2013.04.039
• Publication Date: 2013-04-30T21:05:33Z
• Authors: Xin Zhao; Shicheng Wang; Jinsheng Zhang; Zhiliang Fan; Haibo Min
• Keywords: Maximization; Navigation System

Real-time and accurate fault detection is essential to enhance the aircraft navigation system's reliability safety. The existent methods based on analytical model draws back at simultaneously detecting gradual sudden faults. On account of this reason, we propose an online solution non-analytical model. In article, system established belief rule base (BRB), where measuring residual its changing rate are used as inputs BRB function output. To overcome drawbacks current parameter optimization algorithms for achieve update, a recursive estimation algorithm presented expectation maximization (EM) algorithm. Furthermore, proposed method verified by experiment. Experimental results show that able effectively realize evaluation in output can track state very well, faults be diagnosed real time accurately. addition, ability, especially probability false detection, superior offline method, thus has great improvement.