Sensitivity-based non-linear model predictive control for aircraft descent operations subject to time constraints
The ability to meet a controlled time of arrival while also flying a continuous descent operation will enable environmentally friendly and fuel efficient descent operations while simultaneously maintaining airport throughput. Previous work showed that model predictive control, a guidance strategy ba...
| Autores: | , , |
|---|---|
| Tipo de recurso: | artículo |
| Fecha de publicación: | 2021 |
| País: | España |
| Institución: | Universitat Politècnica de Catalunya (UPC) |
| Repositorio: | UPCommons. Portal del coneixement obert de la UPC |
| Idioma: | inglés |
| OAI Identifier: | oai:upcommons.upc.edu:2117/359503 |
| Acceso en línea: | https://hdl.handle.net/2117/359503 https://dx.doi.org/10.3390/aerospace8120377 |
| Access Level: | acceso abierto |
| Palabra clave: | Navigation (Aeronautics) Trajectory optimization Model predictive control Continuous descent operations Navegació aèria Àrees temàtiques de la UPC::Aeronàutica i espai::Navegació aèria::Procediments operacionals |
| Sumario: | The ability to meet a controlled time of arrival while also flying a continuous descent operation will enable environmentally friendly and fuel efficient descent operations while simultaneously maintaining airport throughput. Previous work showed that model predictive control, a guidance strategy based on a reiterated update of the optimal trajectory during the descent, provides excellent environmental impact mitigation figures while meeting operational constraints in the presence of modeling errors. Despite that, the computational delay associated with the solution of the trajectory optimization problem could lead to performance degradation and stability issues. This paper proposes two guidance strategies based on the theory of neighboring extremals that alleviate this problem. Parametric sensitivities are obtained by linearization of the necessary conditions of optimality along the active optimal trajectory plan to rapidly update it for small perturbations, effectively converting the complex and time consuming non-linear programming problem into a manageable quadratic programming problem. Promising results, derived from more than 4000 simulations, show that the performance of this method is comparable to that of instantaneously recalculating the optimal trajectory at each time sample |
|---|