Application of Rao-Blackwellized particle filtering for estimating remaining useful life of gearbox

We tackle the remaining useful life (RUL) estimating of gearbox problem using conditionally Gaussian state space models and an efficient Monte Carlo method (MCM) known as Rao-Blackwellised particle filtering (RBPF). This paper addresses the problem of estimating the gearbox RUL from the observed degradation data. In this setting, the task of prognosis is to estimate the state probability of operation using the continuous measurements corrupted by Gaussian noise. Data from a full life test for a gearbox are used to validate the proposed methodology; the result fully shows the feasibility and effectiveness of the proposed method.