Projection-based model order reduction for model predictive control of a descending aircraft
Abstract/Contents
- Abstract
- The landing of an aircraft is one of the most important aspects of flight, and also one of the most difficult. This challenge is only amplified in the context of aircraft carrier landings, particularly in adverse flight conditions such as rough seas. Improving the quality of landing motivates the utilization of autonomy and therefore advanced methods for real-time control that account for uncertainty, as this would allow for better and more precise actions to be executed. Unfortunately, current autonomous landing methods are often based on low-fidelity computational models and do not account for non-ideal environmental factors, which limits their applicability. Model predictive control (MPC) methods informed by higher-fidelity computational models that are grounded in computational fluid dynamics (CFD), computational structural dynamics (CSD), and account for fluid-structure interaction (FSI), have a strong potential for significantly improving the technology of autonomous landing in adverse flight conditions. However, materializing this potential requires first addressing the issue that high-fidelity computational models are typically high-dimensional, and therefore impractical for real-time MPC. This thesis rises to the challenge by developing a computational framework for developing a real-time FSI computational technology grounded in CFD, CSD, and projection-based model order reduction. Specifically, the thesis proposes a two-level, data-driven concept for the autonomous landing of aircraft. It features a robust approach for model predictive control; and an innovative, real-time, computational model for FSI and flight dynamics to inform it. The latter is based on the linearization about a pre-designed glideslope trajectory of a high-fidelity, viscous, CFD-CSD-based computational model for flight dynamics; and its projection onto a low-dimensional approximation subspace to achieve real-time performance, while maintaining accuracy. Unlike static lookup tables or regression-based surrogate models based on steady-state wind tunnel data, the proposed real-time computational model allows MPC to be informed by a truly dynamic flight model, rather than a less accurate set of steady-state aerodynamic force and moment data points. While all this work assumes some simplifications, it paves the way for a new approach for achieving autonomous aircraft landing.
Description
| Type of resource | text |
|---|---|
| Form | electronic resource; remote; computer; online resource |
| Extent | 1 online resource. |
| Place | California |
| Place | [Stanford, California] |
| Publisher | [Stanford University] |
| Copyright date | 2021; ©2021 |
| Publication date | 2021; 2021 |
| Issuance | monographic |
| Language | English |
Creators/Contributors
| Author | McClellan, Andrew Richard |
|---|---|
| Degree supervisor | Farhat, Charbel |
| Thesis advisor | Farhat, Charbel |
| Thesis advisor | Alonso, Juan José, 1968- |
| Thesis advisor | Pavone, Marco, 1980- |
| Degree committee member | Alonso, Juan José, 1968- |
| Degree committee member | Pavone, Marco, 1980- |
| Associated with | Stanford University, Department of Aeronautics and Astronautics |
Subjects
| Genre | Theses |
|---|---|
| Genre | Text |
Bibliographic information
| Statement of responsibility | Andrew Richard McClellan. |
|---|---|
| Note | Submitted to the Department of Aeronautics and Astronautics. |
| Thesis | Thesis Ph.D. Stanford University 2021. |
| Location | https://purl.stanford.edu/nh192hk2749 |
Access conditions
- Copyright
- © 2021 by Andrew Richard McClellan
- License
- This work is licensed under a Creative Commons Attribution Non Commercial 3.0 Unported license (CC BY-NC).
Also listed in
Loading usage metrics...