An improved Kalman filter for attitude determination of multi-rotor UAVs based on low-cost MEMS sensors

In this paper, a quaternion based extended Kalman filter (EKF) is developed for determining the attitude and heading of multi-rotor UAVs from the outputs of a self-designed nano Inertial Measurement Unit (nIMU), which consists of low-cost MEMS sensors including a triaxial gyroscope, a triaxial accelerometer and a triaxial magnetometer. In the proposed filtering design, the quaternion associated with the rotation of the quadrotor is included in the state vector with the bias of the sensors. The Gauss-Newton method for parameter optimization is utilized along with Kalman filtering. Preprocessing of sensors' measurements using the Gauss-Newton method yields a computed quaternion input for the Kalman filter, which reduces the dimension of the state vector and linearizes the measurement equations, hence the filter design is significantly simplified. Experimental results have validated the filter design and show good performance of the self-designed nIMU used on quadrotors.