In recent years, with the increase of activities in space, a large number of resident space objects (RSOs) have begun to orbit the Earth. Among them, space debris pose a significant risk to current and future missions that is destined to increase over time due to explosions and collisions phenomena. The scientific community has identified a potential solution in On-Orbit Servicing and Active Debris Removal; thus, the capability to safely conduct proximity operations represents, nowadays, an active area of research. Because of the lack of a priori available information, the task becomes even more challenging when dealing with the most general case of uncooperative and unknown RSOs. With CoMBiNa (COarse Model-Based relatIve NAvigation), a new technique for autonomous inspection of this kind of objects has been proposed in [1]*. By using only stereo-vision measurements, the algorithm combines the advantages of Simultaneous Localization And Mapping (SLAM), a model-based technique called Bayesian Coherent Point Drift (BCPD) and an Unscented Kalman Filter (UKF) to estimate the pose and inertial properties of the target. Nevertheless, CoMBiNa is not expected to converge when dealing with symmetrical RSOs; indeed, due to the unobservable rotations about the symmetry axes, the attitude parameters and angular velocity of the target cannot be properly reconstructed, affecting also the estimation of the other variables of interest. The work proposed in this thesis addresses this problem by suggesting a modified formulation of CoMBiNa to deal specifically with symmetric objects. Notably, this new formulation aims at estimating, together with relative position and velocity between the chaser and the target, the orientation of the symmetry axis of the target body as well as a set of attitude and inertia-coupled quantities. For this purpose, the RSO is approximated to a gyroscopic structure and the fundamental principles of gyroscope theory are exploited. To verify the validity of this approach, a numerical test campaign was performed, which produced promising results. *M. Maestrini. "Satellite Inspection of Unknown Resident Space Objects". PhD thesis, Politecnico di Milano, 2021.
Negli ultimi anni, un elevato numero di oggetti spaziali (conosciuti come RSO – Resident Space Objects) ha iniziato a orbitare attorno alla Terra per via dell’aumento di attività nello spazio. Tra questi, i detriti spaziali rappresentano un rischio significativo che, a causa di potenziali esplosioni e collisioni, è destinato ad aumentare nel tempo. La comunità scientifica ha identificato nel "Servizio in Orbita" (On-Orbit Servicing) e nella rimozione attiva dei detriti una possibile soluzione a tale problema; pertanto, la capacità di condurre in sicurezza operazioni di prossimità rappresenta ad oggi un’area attiva di ricerca. Tale compito diventa ancora più arduo quando si ha a che fare con RSO non cooperativi e sconosciuti, per via della mancanza di informazioni disponibili a priori. CoMBiNa (COarse Model-Based relatIve NAvigation) rappresenta una nuova tecnica per l’ispezione autonoma di questo tipo di oggetti, proposta in [1]*. Utilizzando solamente misure ottenute tramite una stereo camera, l’algoritmo combina i vantaggi della localizzazione e mappatura simultanea (SLAM), una tecnica model-based chiamata Bayesian Coherent Point Drift (BCPD) e un Filtro di Kalman Unscented (UKF) per stimare la posa e le proprietà inerziali dell’oggetto target. Tuttavia, CoMBiNa presenta problemi di convergenza quando applicato a RSO simmetrici; infatti, a causa delle rotazioni non osservabili attorno agli assi di simmetria, l’assetto del target non viene ricostruito correttamente, influenzando anche la convergenza delle altre variabili di interesse. Il lavoro proposto in questa tesi affronta il problema proponendo una formulazione modificata di CoMBiNa in grado di trattare specificatamente oggetti simmetrici. In particolare, questa nuova formulazione mira a stimare, insieme alla posizione e velocità relativa tra il satellite attivo e il target passivo, l’orientamento dell’asse di simmetria del target e, inoltre, un set di variabili che mettono in relazione il suo assetto e le sue proprietà inerziali. A tal fine, l’RSO è approssimato a una struttura giroscopica in modo da sfruttare i principi fondamentali della teoria dei giroscopi. Per verificare la validità di tale approccio è stata eseguita una simulazione numerica che ha prodotto risultati promettenti. *M. Maestrini. "Satellite Inspection of Unknown Resident Space Objects". PhD thesis, Politecnico di Milano, 2021.
Relative navigation at unknown uncooperative symmetric objects
De Luca, Maria Alessandra
2021/2022
Abstract
In recent years, with the increase of activities in space, a large number of resident space objects (RSOs) have begun to orbit the Earth. Among them, space debris pose a significant risk to current and future missions that is destined to increase over time due to explosions and collisions phenomena. The scientific community has identified a potential solution in On-Orbit Servicing and Active Debris Removal; thus, the capability to safely conduct proximity operations represents, nowadays, an active area of research. Because of the lack of a priori available information, the task becomes even more challenging when dealing with the most general case of uncooperative and unknown RSOs. With CoMBiNa (COarse Model-Based relatIve NAvigation), a new technique for autonomous inspection of this kind of objects has been proposed in [1]*. By using only stereo-vision measurements, the algorithm combines the advantages of Simultaneous Localization And Mapping (SLAM), a model-based technique called Bayesian Coherent Point Drift (BCPD) and an Unscented Kalman Filter (UKF) to estimate the pose and inertial properties of the target. Nevertheless, CoMBiNa is not expected to converge when dealing with symmetrical RSOs; indeed, due to the unobservable rotations about the symmetry axes, the attitude parameters and angular velocity of the target cannot be properly reconstructed, affecting also the estimation of the other variables of interest. The work proposed in this thesis addresses this problem by suggesting a modified formulation of CoMBiNa to deal specifically with symmetric objects. Notably, this new formulation aims at estimating, together with relative position and velocity between the chaser and the target, the orientation of the symmetry axis of the target body as well as a set of attitude and inertia-coupled quantities. For this purpose, the RSO is approximated to a gyroscopic structure and the fundamental principles of gyroscope theory are exploited. To verify the validity of this approach, a numerical test campaign was performed, which produced promising results. *M. Maestrini. "Satellite Inspection of Unknown Resident Space Objects". PhD thesis, Politecnico di Milano, 2021.File | Dimensione | Formato | |
---|---|---|---|
Thesis_DeLuca.pdf
accessibile in internet solo dagli utenti autorizzati
Descrizione: Thesis
Dimensione
5.22 MB
Formato
Adobe PDF
|
5.22 MB | Adobe PDF | Visualizza/Apri |
Executive_Summary_Thesis_DeLuca.pdf
accessibile in internet solo dagli utenti autorizzati
Descrizione: Executive Summary
Dimensione
3.48 MB
Formato
Adobe PDF
|
3.48 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/10589/195621