Semi-Aerodynamic Model Aided Invariant Kalman Filtering for UAV Full-State Estimation

TL;DR

This paper introduces a semi-aerodynamic model fused with invariant Kalman filtering for fixed-wing UAVs, improving real-time full-state estimation accuracy and robustness, especially in GNSS-denied environments, using deep learning for angle prediction.

Contribution

It develops a novel ES-RIEKF-based data fusion framework combined with a semi-aerodynamic model and LSTM-based angle prediction, enhancing UAV navigation accuracy and robustness.

Findings

Achieved real-time pose and wind estimation with high accuracy.

Demonstrated robustness with positioning errors within 30 meters during GNSS denial.

Validated improvements over traditional filters using real UAV flight data.

Abstract

Due to the state trajectory-independent features of invariant Kalman filtering (InEKF), it has attracted widespread attention in the research community for its significantly improved state estimation accuracy and convergence under disturbance. In this paper, we formulate the full-source data fusion navigation problem for fixed-wing unmanned aerial vehicle (UAV) within a framework based on error state right-invariant extended Kalman filtering (ES-RIEKF) on Lie groups. We merge measurements from a multi-rate onboard sensor network on UAVs to achieve real-time estimation of pose, air flow angles, and wind speed. Detailed derivations are provided, and the algorithm's convergence and accuracy improvements over established methods like Error State EKF (ES-EKF) and Nonlinear Complementary Filter (NCF) are demonstrated using real-flight data from UAVs. Additionally, we introduce a semi-aerodynamic model fusion framework that relies solely on ground-measurable parameters. We design and train an Long Short Term Memory (LSTM) deep network to achieve drift-free prediction of the UAV's angle of attack (AOA) and side-slip angle (SA) using easily obtainable onboard data like control surface deflections, thereby significantly reducing dependency on GNSS or complicated aerodynamic model parameters. Further, we validate the algorithm's robust advantages under GNSS denied, where flight data shows that the maximum positioning error stays within 30 meters over a 130-second denial period. To the best of our knowledge, this study is the first to apply ES-RIEKF to full-source navigation applications for fixed-wing UAVs, aiming to provide engineering references for designers. Our implementations using MATLAB/Simulink will open source.