Fault detection of Integrated navigation systems based on one-class support vector machine

The fault of integrated navigation system will affect the precision of navigation result and even the whole system, so the fault detection of integrated navigation is very important. However, the data of the integrated navigation system is typical skew data, that is, the amount of correct data is much more than the amount of faulty data, resulting in a low fault detection rate. To solve this problem, this paper proposes a fault detection method based on one-class support vector machine(OCSVM). This method fully learns and utilizes the data feature information of correct data and constructs a classifier to distinguish data types. Simulation results show that OCSVM has improved detection rate and false alarm rate compared with standard SVM.