An EKF Based on Aerodynamic Constraints for Fixed-Wing AAV Attitude Estimation

Abstract

Low-cost inertial measurement units (IMUs) are commonly utilized for estimating the attitude of autonomous aerial vehicles (AAVs). However, accelerometer readings from AAVs in flight encompass both gravitational and external accelerations (EAs). The presence of EAs impacts the accuracy of attitude estimation. To enhance AAVs attitude estimation accuracy, this article proposes a two-stage extended Kalman filter (EKF) that incorporates constraints from aerodynamic forces and gravitational acceleration. In the first stage, an EA model based on aerodynamic force constraints is developed to calculate the y- and z-axis components of gravitational acceleration. Since low-cost AAVs do not have thrust sensors installed, the second stage employs gravitational acceleration constraints to compute the x-axis component of gravitational acceleration. After acquiring the gravitational acceleration, optimal attitude estimation can be performed. To validate the performance of the proposed algorithm, a fixed-wing AAV experimental platform was established and subjected to flight-testing. The results indicate that the proposed algorithm achieves greater estimation accuracy compared to alternative algorithms during extended flights and high-dynamic or large-maneuvering conditions.