Alain Yetendje

Brief paper: Multisensor fusion fault tolerant control

In this paper, a multisensor fusion fault tolerant control system with fault detection and identification via set separation is presented. The fault detection and identification unit verifies that for each sensor-estimator combination, the estimation tracking errors lie inside pre-computed sets and discards faulty sensors when their associated estimation tracking errors leave the sets. An active fault tolerant controller is obtained, where the remaining healthy estimates are combined using a technique based on the optimal fusion criterion in the linear minimum-variance sense. The fused estimates are then used to implement a state feedback tracking controller. We ensure closed-loop stability and performance under the occurrence of abrupt sensor faults. Experimental validation, illustrating the multisensor fusion fault tolerant control strategy is included.

Robust MPC design for fault tolerance of constrained multisensor linear systems

In this paper, a robust fault-tolerant control strategy for constrained multisensor linear systems, subject to sensor faults and in the presence of bounded state and output disturbances is proposed. The scheme verifies that for each sensors-estimator combination, suitable residual variables lie inside pre-computed sets and selects the more appropriate combination based on a chosen criterion. An active fault tolerant output feedback controller yields an MPC-based control law and by means of the notion of “tube” of trajectories, we ensure robust closed-loop exponential stability and good performance in the fault-free case and under the occurrence of abrupt sensor faults.

Multisensor fusion fault-tolerant control with diagnosis via a set separation principle

In this paper, a multisensor fusion fault-tolerant control system with fault detection and isolation via set separation is presented. The fault detection and isolation unit verifies that for each sensor-estimator combination, the estimation tracking errors lie inside pre-computed sets and discards faulty sensors when their associated estimation tracking errors leave the sets. An active fault tolerant controller is obtained, where the remaining healthy estimates are combined using a technique based on the optimal fusion criterion in the linear minimum-variance sense, recently proposed in the literature. The fused estimates are then used to implement a state feedback tracking controller. We ensure closed-loop stability and good performance under the occurrence of abrupt sensor faults.

Robust MPC Multicontroller Design for Actuator Fault Tolerance of Constrained Systems

Abstract In this paper, we present a robust actuator fault-tolerant control strategy for constrained linear systems, in the presence of bounded state and output disturbances. The scheme is based on a bank of state estimators which match different fault situations that can occur in the system. A fault detection and isolation (FDI) unit verifies that suitable residual variables lie inside pre-computed sets and selects the estimate that matches the current plant behaviour. A bank of robustly stabilising tube-based MPC control laws is designed, each associated to a fault scenario and the appropriate controller is selected among them using the information provided by the FDI module. By means of “tubes” of trajectories, we ensure robust closed-loop exponential stability of the constrained system and good performance in the fault-free case and under the occurrence of abrupt actuator faults.

Fault-tolerant control of a magnetic levitation system using virtual-sensor-based reconfiguration

In this paper, a fault tolerant control strategy, which combines a fault detection and identification (FDI) method based on an invariant-set approach with controller reconfiguration based on the use of a virtual sensor, is implemented on a magnetic levitation (MAGLEV) system. The MAGLEV system includes two sensors which, together with a nominal observer-based feedback controller with integral action, are used to stabilise a steel ball at a desired position in the air. The FDI unit employs two observers that can detect the faulty and healthy situations based on a “set-separation” approach. The closed-loop system is reconfigured by means of a virtual sensor which is adapted to the fault situation detected by the FDI unit. Experimental results on a laboratory apparatus confirm the effectiveness of the proposed strategy.

Fault-Tolerant Switching Control of a Magnetic Levitation Syste

Abstract In this paper, a fault tolerant switching control strategy is implemented on a magnetic levitation (MAGLEV) system. Two sensors are embedded in the MAGLEV system and their measurements used by two independent estimators. Each sensors-estimator combination, together with a feedback controller can levitate and stabilise a one-inch steel ball at a desired position in the air. The paper focuses on the design and testing of a switching scheme which, at each instant of time, selects the sensors-estimator combination that provides the best closed loop performance based on a chosen criterion. Theoretical results on the system linearisation around an operating point ensure local closed-loop stability and good performance under the occurrence of an abrupt fault in one of the plant sensors. Experimental results are provided which confirm the fault tolerant capabilities of the strategy.

Multisensor fusion fault-tolerant control of a magnetic levitation system

In this paper, a multisensor fusion fault-tolerant control system with fault detection and isolation via set separation is implemented on a magnetic levitation (MAGLEV) system. Three sensors are embedded in the MAGLEV system and their measurements used by three independent estimators. A fault detection and isolation unit verifies that for each sensor-estimator combination, the estimation tracking errors lie inside pre-computed sets and discards faulty sensors when their associated estimation tracking errors leave the sets. An active fault tolerant controller is obtained, where the remaining healthy estimates are combined using an optimal fusion criterion. Theoretical results on the system linearisation around an operating point ensure local closed-loop stability and good performance under the occurrence of abrupt sensor faults. Experimental results are provided which confirm the fault tolerant capabilities of the strategy.