Attitude estimation for UAV with low-cost IMU/ADS based on adaptive-gain complementary filter

Abstract

This paper describes a robust and practical complementary filter (CF) algorithm for unmanned aerial vehicle (UAV) attitude estimation with low-cost inertial measurement unit (IMU) and embedded air data system (ADS). Utilizing a fuzzy logical system, the UAV dynamic modes including different accelerations and turns can be attained. Based on the compensation of acceleration and centrifugal forces in turns using ADS information, the gains of complementary filter adapts to the dynamic modes to yield robust performances. The simulation and experimental results show that the proposed adaptive-gain complementary filter approach can obtain robust and accurate attitude estimation even when the UAV is subject to strong acceleration or in turn mode.