Present article provides a set-based fault tolerant control strategy for multi-sensor systems, where sensors are communicating with a controller via a shared network. Possible faults, such as abrupt sensor outages and network-induced delays, are identified as degradation modes which might affect the information provided by each sensor. Measurements that are transmitted from a sensor to the controller are characterized by a residual signal which is sensitive to the sensor's abrupt faults and network-induced delays. In order to avoid control based on information which is provided by a faulty sensor, we designed a fault detection and isolation mechanism that is implemented through a set membership evaluation. This evaluation differentiates between ``healthy'', ``faulty'' and ``delayed'' data transmission. Unequivocal fault detection and isolation are assured if the corresponding sets are disjoint. Since in general this is not the case, sets separation is enforced by a reference governor. Fault detection and isolation mechanism is design in order to transmit only measurements from sensors which are fully operational, even if potentially affected by delays. If there is a delayed information that reaches the controller, then control action is reconfigured in order to govern the plant as close as possible to the reference signal. Such control action is provided by a model-based controller with compensation block. Sufficient condition that guarantees the existence of the compensation signal is presented as well.
The present paper deals with fault tolerant control for linear dynamics with additive disturbances. The control action is generated based on information collected from a redundant, multi-sensors network. Delays that may appear during plant measurements transmission through real communication channels are considered as faults. Depending on presence of delay in feedback loop, dierent invariant sets can be computed. We show that fault tolerant control can be achieved through invariant sets separation with respect to dierent delay values. Sets separation is accomplished for specic values of the reference signal. Therefore, we introduce in the loop a reference governor which is designed by a receding horizon technique. Thus, we provide reference signals which practically guarantee fault detection and identication in real time.