An Online Fault Diagnosis Method for Actuators of Quadrotor UAV with Novel Configuration Based on IMM

This paper studies the problem of actuator fault of a new type of quadrotor UAV with high payload and energy utilization efficiency. When the UAV is performing tasks in a complex and dangerous environment, it is extremely prone to damage to the actuator structure, which will reduce the efficiency of the actuator, resulting in loss of control or even a crash. Quadrotor UAV is a non-linear strong coupling system and there are many rotor actuators, so the fault situation is complex. In the paper, based on the idea of IMM, establishes a typical fault model set and monitors each rotor actuator fault situation online through multiple Kalman filters based on fault models. Firstly, considering the redundancy of the power unit, combined with the characteristics of the model of the new configuration quadrotor UAV to detect and locate the fault, then calculate the weight coefficient and model matching rate of each fault model, and finally perform weighted fusion and estimate the efficiency parameters of the fault actuator. The simulation is carried out using the PID controller. The results show that the method can quickly detect and locate the fault of a single actuator about the new quadrotor UAV, and can estimate the efficiency parameters of the rotor actuator after the fault.