Intelligent Multiple-Fault Diagnosis of a Mobile Robot System in the Presence of Hide Effect

This work focuses on the development of a method to multiple-fault diagnosis of a mobile robot system in the presence of hide effect. The faults under consideration include parametric faults, sensor faults as well as fault modes. Multiple faults can happen at different time instances, and fault signature of the earlier fault can hide fault signature of the latter one which adds to difficulty of the latter fault diagnosis. In order to overcome this problem, a method for the latter fault isolation is introduced by using the second level residual (SLR). After fault candidates are obtained using the SLR, a fuzzy harmony search algorithm is proposed for fault identification. Simulation results validate the effectiveness of the proposed method.