A fault detection structure based on the federated Kalman filter is proposed. This structure uses the residual between the common states of each local filter and the reference filter for fault detection. Two fault detection algorithms are proposed: the χ2 test method and the Elman neural network test method. A simulation study is carried out using the integrated navigation system as an example. Compared with other algorithms, this algorithm is simpler and more reliable in calculation. It can not only quickly detect external sensor and reference system faults, but also has good fault tolerance performance. It can quickly detect faults and isolate them, so that the fusion system still maintains a high accuracy. Keywords: fault detection; filter; neural network; integrated navigation
You Might Like
Recommended ContentMore
Open source project More
Popular Components
Searched by Users
Just Take a LookMore
Trending Downloads
Trending ArticlesMore