MPC-based FTC with FDD against actuator faults of UAVs

Abstract

The increasing development of unmanned aerial vehicles (UAVs) and the requirements of high autonomy and safety levels require that the controller of UAVs should possess fault-tolerant/reconfigurable function to accommodate unpredicted situations such as actuator faults, sensor faults, or aircraft damage. This paper addresses the partial loss of control effectiveness (LOE) of actuators in a quadrotor UAV using model predictive control (MPC) with terminal constraints, which allows to track a reference command even if in the presence of actuator faults. The proposed fault-tolerant control system (FTCS) adopts MPC technique to design fault-tolerant controller and state-augmented Kalman filter (SAKF) to achieve fault detection and diagnosis (FDD) function. Simulation results based on a quadrotor UAV demonstrate that the proposed fault-tolerant controller has a good performance in accommodating actuator faults.