Performance of a Flush Airdata Sensor in a Particle Filter-Based Re-entry Navigation System

Master thesis (2017)

Authors

Contributors

E. Mooij (mentor)
Programme
Astrodynamics & Space Missions (Aerospace Engineering) (TU Delft)
Copyright: © 2017 Rijnsdorp, J.
Navigation Particle Filter Space Extended Kalman Filter Unscented Kalman Filter Re-entry EKF UKF Flush Airdata Sensor GNC-system
More Info
expand_more

Published Date

10-05-2017

Reuse Rights

Other than for strictly personal use, it is not permitted to download, forward or distribute the text or part of it, without the consent of the author(s) and/or copyright holder(s), unless the work is under an open content license such as Creative Commons.

Programme

Astrodynamics & Space Missions

Abstract

Re-entry is a term that is applied when one would like to transport some (human) payload from an orbit around a certain body towards the surface of that body. With an entry velocity of multiple kilometers per second, having the vehicle carrying this payload not smashed into pieces at the end of flight is one of the key points of re-entry. Using a guidance, navigation, and control system (GNC), one is capable of defining a path through the atmosphere towards a predefined target, and following that trajectory so that a safe landing can be assured. This thesis focuses on the navigation module of the GNC system, where the application of a Flush Airdata Sensor (FADS) and the comparison between the Extended Kalman Filter (EKF), the Unscented Kalman Filter (UKF), and the Particle Filter (PF) is considered. The approach that is taken to evaluate the performance of the three filters and the FADS is based on the design of a re-entry simulator, which provides a 6 degree-of-freedom simulated motion through the atmosphere. The guidance system consists of a set of commanded attitude angles and for the control system, a Linear Quadratic Regulator (LQR) is designed that is able to control the vehicle throughout the hypersonic descent of the atmospheric re-entry flight. Then, the EKF, UKF, and PF are designed and evaluated individually, using measurements from an Inertial Measurement Unit (IMU), a Global Positioning System (GPS) receiver and the FADS. From the individual assessment of the filters, it is found that the PF outperforms the two Kalman filters. Concluding this, the PF-based navigation module is integrated into the simulator, thereby completing the GNC-system. The FADS shows improvement of the position estimation results, as the final error for a navigation module without the FADS is 0.93 m for the altitude, whereas the integration of the FADS in that module results in a final estimation error of 0.62 m. In terms of velocity, errors in the magnitude of ~0.02 m/s are found, and, for the attitude angles, errors of ~0.02 degrees are estimated.