Attitude Estimation of Miniature Unmanned Helicopter using Unscented Kalman Filter

Attitude Estimation is critical for autonomous flight of aircraft. In this paper, an attitude estimation method for Miniature Unmanned Helicopter (MUH) is presented. The method is based on Unscented Kalman Filter (UKF). It employs Quaternion to represent rotary motion. It utilizes information from three low-cost single-axis MEMS gyroscopes, two dual-axis accelerometers and a three-axis electronic compass. System state equations are established on quaternion vector and gyro bias vector while measurement equations are based on measured attitude angles. The method fuses information from several different kinds of sensors to get an optimal result. Simulations and real flight experiments are carried out to verify the validity of the method. The results show that the method overcomes the drawbacks of each sensor and is suitable for application on MUH. Compared to attitude estimation method using Extended Kalman Filter (EKF), the method avoids the calculation of Jacobian matrices.