An attitude estimation system for the unmanned helicopter is presented.The system is based on an inertial measurement unit（IMU）.By using Allan variance analysis method

the inertial sensor’s models are built up.The extended Kalman filtering（EKF） algorithm and the complementary filtering algorithm are drawn into flight attitude determination for a helicopter.The methodologies and implementation of the two algorithms are well elaborated.The experimental tests have verified that both Kalman filtering and complementary filtering can obtain a very effective attitude estimation system

and for an unmanned helicopter

the complementary filter can yield better performance than EKF.