In recent years, the interest in adaptive controllers has grown a lot thanks to the great advantages that can derive from their use and to the potential vastness of the field of use of this type of controls. The aviation industry can effectively use these controllers on all types of aircraft and Unmanned Aerial Vehicle (UAVs) to improve the flight comfort and safety thanks to the rapid suppression of external disturbances and the ability to counteract on-board failures such as the loss of an engine, as well as to significantly reduce the costs deriving from re-calibration of the Flight Control Systems (FCSs) parameters which in many cases is required to restore the optimal behaviour of the aircraft in the event of subsequent modifications. These advantages, however, clash with the great difficulty in the verification and validation (V&V) phase of these controls due to the difficulty of obtaining guarantees on the global stability of these systems, which is a mandatory requirement for aircraft and in general for all the safety-critical systems. The purpose of this thesis is to investigate and address the problems related to the implementation of the Model Reference Adaptive Controllers (MRACs) and the L1 Adaptive Controller (L1-AC) in the outer-loops of the position-attitude cascade controllers, which represent the reference architecture for the PID controllers in aeronautical industry. The problem of the inner-loop bandwidth influence on the outer one has been addressed proposing a way to evaluate such influence, implementing a highperformance MRAC with hedge signal for the light-weight high-manoeuvrable quadrotor ANT-X, simulating its response to an 8-shape input trajectory on the simulator and, finally, comparing its response with that of the L1-AC. A high-performance adaptive controller was then obtained which improves the response of the system compared to the L1-AC, in the presence of an external disturbance given by the unmodeled aerodynamic drag.
Negli ultimi anni l’interesse per i controllori adattivi `e cresciuto molto grazie ai grandi vantaggi che possono derivare dal loro utilizzo e alla potenziale vastit`a del campo di applicazione di questo tipo di controlli. L’industria aeronautica pu`o utilizzare efficacemente questi controllori su tutti i tipi di aeromobili e droni a pilotaggio remoto (UAV) per migliorare il comfort e la sicurezza di volo grazie alla rapida soppressione dei disturbi esterni e alla capacit`a di contrastare le avarie a bordo come la perdita di un motore, nonch`e per ridurre sensibilmente i costi derivanti dalla ritaratura dei parametri dei Flight Control Systems (FCSs) che in molti casi `e richiesta per ripristinare il comportamento ottimale della macchina in caso di modifiche successive. Tali vantaggi si scontrano per`o con la grande difficolt` a in fase di verifica e validazione (V&V) di tali controlli a causa della difficolt`a di ottenere garanzie sulla stabilit`a globale di tali sistemi, requisito obbligatorio per gli aeromobili e in generale per tutti i sistemi critici per la sicurezza. Lo scopo di questa tesi `e quello di indagare e affrontare i problemi relativi all’implementazione del Model Reference Adaptive Controllers (MRACs) e del L1 Adaptive Controller (L1-AC) negli anelli esterni dei controllori in cascata posizione-assetto, che rappresentano l’architettura di riferimento per i controllori PID nell’ industria aeronautica. `E stato affrontato il problema dell’influenza della larghezza di banda dell’anello interno su quello esterno proponendo un modo per valutare tale influenza, implementando un MRAC ad alte prestazioni con hedge signal per il quadrirotore leggero altamente manovrabile ANT-X, simulando la sua risposta a una traiettoria in ingresso a forma di 8 sul simulatore e, infine, confrontando la sua risposta con quella del L1-AC. `E stato quindi ottenuto un controllore adattativo ad alte prestazioni che migliora la risposta del sistema rispetto al L1-AC, in presenza di un disturbo esterno dato dalla resistenza aerodinamica non modellizzata.
Adaptive control for position-attitude cascade controller for UAVs
Della Libera, Alberto
2020/2021
Abstract
In recent years, the interest in adaptive controllers has grown a lot thanks to the great advantages that can derive from their use and to the potential vastness of the field of use of this type of controls. The aviation industry can effectively use these controllers on all types of aircraft and Unmanned Aerial Vehicle (UAVs) to improve the flight comfort and safety thanks to the rapid suppression of external disturbances and the ability to counteract on-board failures such as the loss of an engine, as well as to significantly reduce the costs deriving from re-calibration of the Flight Control Systems (FCSs) parameters which in many cases is required to restore the optimal behaviour of the aircraft in the event of subsequent modifications. These advantages, however, clash with the great difficulty in the verification and validation (V&V) phase of these controls due to the difficulty of obtaining guarantees on the global stability of these systems, which is a mandatory requirement for aircraft and in general for all the safety-critical systems. The purpose of this thesis is to investigate and address the problems related to the implementation of the Model Reference Adaptive Controllers (MRACs) and the L1 Adaptive Controller (L1-AC) in the outer-loops of the position-attitude cascade controllers, which represent the reference architecture for the PID controllers in aeronautical industry. The problem of the inner-loop bandwidth influence on the outer one has been addressed proposing a way to evaluate such influence, implementing a highperformance MRAC with hedge signal for the light-weight high-manoeuvrable quadrotor ANT-X, simulating its response to an 8-shape input trajectory on the simulator and, finally, comparing its response with that of the L1-AC. A high-performance adaptive controller was then obtained which improves the response of the system compared to the L1-AC, in the presence of an external disturbance given by the unmodeled aerodynamic drag.File | Dimensione | Formato | |
---|---|---|---|
898408_DELLA_LIBERA.pdf
Open Access dal 30/11/2024
Descrizione: Adaptive control for position-attitude cascade controller for UAVs
Dimensione
6.67 MB
Formato
Adobe PDF
|
6.67 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/10589/182473