View article

In this contribution, a method for implementation of AWBT compensation, when the residual signal can not be measured, is presented. The approach consists into estimate the residual signal from a FDI filter. In this case, the residual signal is considered as fault. Fault detection and isolation conditions are applied in order to determine what actuator is saturated. Then, the FDI filter is designed to separate the residual signals represented as faults. Based-on passive and active fault tolerant control, two methods for the implantation of compensated controllers are also presented. For both cases, the saturation measurement problem or the use of actuators models is avoided. The compensation is guaranteed, spite of the changes in the saturation levels; this is, it is not necessary to specify the saturation levels a priori.