Flight control systems for high-performance aircraft and drones must manage significant nonlinearities and uncertainties to meet certification standards. Incremental Nonlinear Dynamic Inversion (INDI) is a promising control framework, yet its sensitivity to time delays, sensor dynamics, and modeling errors has hindered its industrial adoption. This thesis is a step forward in overcoming these limitations by developing a novel adaptive INDI architecture that enables systematic and certifiable design. The main innovation is the use of an Extended State Observer that treats system non-idealities as a single, aggregated disturbance. With respect to L1 adaptive control, which was already used for the same problem, the proposed solution is more transparent to the control designer and offers more degrees of freedom in recovering the stability margins of the baseline control system. Validation on a longitudinal autopilot demonstrates superior robustness to underestimated time delays and sensor phase lag compared to other adaptive architectures, maintaining stability in highly perturbed scenarios where baseline INDI controllers fail. By transforming a complex manual process into an automated robust synthesis, this work contributes to streamline the adoption of INDI in real aerospace applications, transitioning it from an experimental technique into a potentially certifiable solution.
I sistemi di controllo per velivoli ad alte prestazioni e droni devono gestire forti non linearità e significative incertezze per soddisfare i requisiti di certificazione. L’Inversione Dinamica Nonlineare Incrementale (INDI) rappresenta un’architettura promettente, ma la sua fragilità rispetto a ritardi temporali, dinamiche sensoriali ed errori di modellazione ne ha finora ostacolato l’adozione industriale. Questa tesi compie un passo importante verso il superamento di tali limiti, introducendo un’estensione adattiva dell’INDI capace di garantire una progettazione sistematica e potenzialmente certificabile. Il fulcro dell’architettura è un Osservatore di Stato Esteso (ESO) che incorpora in un unico termine perturbativo tutte le non idealità del sistema. Rispetto al controllo adattivo di tipo L1, già applicato allo stesso problema, l’approccio proposto offre maggiore trasparenza progettuale e più gradi di libertà per il recupero dei margini di stabilità. La validazione su un autopilota per il controllo longitudinale dimostra robustezza nei confronti di ritardi di tempo sottostimati e perdite di fase indotte dai sensori, assicurando stabilità anche in scenari fortemente perturbati dove i controllori INDI convenzionali falliscono. Trasformando un complesso processo manuale in una sintesi robusta, trasparente e automatizzata, questo lavoro semplifica l’adozione dell’INDI in applicazioni aerospaziali reali.
Adaptive augmentation of Incremental NDI laws with systematic H-infinity tuning
FRANCESCHINI, GIULIO
2024/2025
Abstract
Flight control systems for high-performance aircraft and drones must manage significant nonlinearities and uncertainties to meet certification standards. Incremental Nonlinear Dynamic Inversion (INDI) is a promising control framework, yet its sensitivity to time delays, sensor dynamics, and modeling errors has hindered its industrial adoption. This thesis is a step forward in overcoming these limitations by developing a novel adaptive INDI architecture that enables systematic and certifiable design. The main innovation is the use of an Extended State Observer that treats system non-idealities as a single, aggregated disturbance. With respect to L1 adaptive control, which was already used for the same problem, the proposed solution is more transparent to the control designer and offers more degrees of freedom in recovering the stability margins of the baseline control system. Validation on a longitudinal autopilot demonstrates superior robustness to underestimated time delays and sensor phase lag compared to other adaptive architectures, maintaining stability in highly perturbed scenarios where baseline INDI controllers fail. By transforming a complex manual process into an automated robust synthesis, this work contributes to streamline the adoption of INDI in real aerospace applications, transitioning it from an experimental technique into a potentially certifiable solution.File | Dimensione | Formato | |
---|---|---|---|
2025_07_Franceschini_Tesi_01.pdf
accessibile in internet per tutti
Descrizione: Testo della tesi
Dimensione
4.31 MB
Formato
Adobe PDF
|
4.31 MB | Adobe PDF | Visualizza/Apri |
2025_07_Franceschini_Executive Summary_02.pdf
accessibile in internet per tutti
Descrizione: Testo dell'executive summary
Dimensione
2.35 MB
Formato
Adobe PDF
|
2.35 MB | Adobe PDF | Visualizza/Apri |
I documenti in POLITesi sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
https://hdl.handle.net/10589/240519