IMU Sensor Fault Diagnosis and Estimation for Quadrotor UAVs

This paper presents a fault detection, isolation and estimation scheme for sensor bias fault in inertial measurement units (IMUs) of quadrotor unmanned air vehicles (UAVs). By using estimated roll and pitch angles obtained using a sliding-mode observer method, a diagnostic algorithm is developed for detecting, isolating and estimating sensor bias fault in the gyroscope and accelerometer measurements. The stability and estimation performance properties of the fault estimation algorithm are rigorously established. Flight data collected from a real-time quadrotor test environment is used to illustrate the effectiveness of the proposed method.