Fault Detection, Diagnosis and Fault Tolerant Output Control for a Remotely Operated Vehicle

Abstract

An active fault tolerant control scheme for an underwater remotely operated vehicle is proposed in this work when only position and heading measurements are available. The diagnosis technique is based on a second order sliding mode observer, allowing to estimate both the unmeasured system state, namely linear and angular speeds, and the fault extent. The fault estimation is fed to a thrust allocation algorithm which achieves fault tolerant capabilities independently of the controller, attaining robustness with respect to actuator faults with any suitable control law. The overall fault diagnosis and fault tolerant scheme has been tested in simulation scenario using four different controllers and using the real world parameters of the SNAMprogetti remotely operated vehicle, thus validating the effectiveness of the proposed solution.