Fault forecast method based on particle filter

Fault prediction based on particle filter approach is designed for dead reckoning investigation hybrid system, it utilizes a group of weighted particles to evaluate the system state, meanwhile, the fault state distribution and fault probability density distribution are calculated. Therefore, we can predict the fault probability and the fault type, furthermore, the broken-down time step can be assessed. The experimental results show that fault prediction based on particle filter can estimate the fault type for dead reckoning investigation hybrid system effectively.