Rotorcraft UAV Actuator Failure Detection based on a New Adaptive set-membership Filter

Actuator failure detection method based on a new Adaptive Extended Set-Membership Filter (AESMF) is proposed for Rotorcraft Unmanned Aerial Vehicle (RUAV). The AEMSF proposed in this paper is based on MIT method to optimize the set boundaries of process noises which may be incorrect in modeling or time-variant in operation; estimation stability and boundaries accuracy can be improved compared to the conventional ESMF. Actuator Healthy Coefficients (AHCs) is introduced into the dynamics of RUAV to denote the actuator failure model. Based on AESMF, online estimation of the AHCs can be obtained along with the flight state. With the estimated AHCs, actuator failure can be detected as soon as possible which provide valuable information for fault tolerant control. Efficiency and improvement of this method compared with other online parameters estimation methods is demonstrated by simulation using ServoHeli-20 model