The problem of controlling the trajectory of aerospace vehicles from an arbitrary initial position and initial velocity to a desired target position with constrained, free or pointed terminal velocity in a specific time is of fundamental interest as an optimal control problem. The purpose of this thesis is to design an optimized closed loop solution in low-thrust, long duration orbit transfer environment considering two different scenarios – Earth to Mars and GTO to GEO. Low-thrust transfers are taken into account because less fuel is required when using electric thrusters which results in less launch mass and less cost. The zero-effort-miss/zero-effort-velocity feedback guidance algorithm which has great potential for autonomous onboard implementation is applied. Unfortunately it does not give an optimal solution because during the entire mission some assumptions are violated. A way to improve the performances is to divide the total flight time into more segments, determining with a rigorous optimization method near-optimal waypoints to connect the different segments. Open loop energy-fuel-optimal trajectories generated by L.Ferrella and F.Topputo [1] are considered as reference where some arbitrary points are put to be targeted by ZEMZEV guidance. An accurate study is developed valuating performances which vary with the number of waypoints. Afterwards an optimization problem, parametrized with the position of the points is set in order to determine the minimum set of waypoints necessary for close-to-fuel-optimal waypoint space guidance. Improved results of optimization are figured out and compared with the random positioning reaching the conclusion that way-point guidance scheme is able to combine open-loop optimal guidance with ZEM/ZEV targeting capabilities. The same guidance is applied also to challenging cases – i.e. cases with a very low limit thrust which results in controlling a trajectory with more than 500 spirals. Finally the model is simulated in a more realistic scenario to verify the robustness of the system performing Monte Carlo analysis. By employing the zero-effort-miss/zero-effort-velocity algorithm in the highly nonlinear orbital transfer and comparing with corresponding open-loop optimal solutions, its near-optimality is further verified. Furthermore the algorithm can even compete with corresponding open-loop optimal solutions, while its feedback characteristics make it more suitable to deal with uncertainties and perturbations.
Il problema del controllo di traiettorie compiute da veicoli aerospaziali da una posizione e velocità iniziale arbitraria a una desiderata posizione finale è di fondamentale interesse come problema di controllo ottimo. L’obiettivo di questa tesi è di progettare una soluzione ottima in anello chiuso nell’ambito di trasferimenti orbitali di lunga durata con bassa spinta, considerando due diversi scenari (Terra-Marte e GTO-GEO). Sono presi in considerazione i trasferimenti a bassa spinta in quanto i motori a propulsione elettrica utilizzati richiedono una quantità di propellente minore e quindi minore massa al lancio e minori costi. E’ applicato un algoritmo di guida ZEMZEV che può essere facilmente implementato autonomamente a bordo. Quest’ultimo purtroppo non fornisce una soluzione ottima poiché durante l’intera missione alcune ipotesi sono violate. Un modo per migliorare le prestazioni è dividere il tempo totale di volo in più tratti, determinando un certo numero di waypoints con un metodo di ottimizzazione rigoroso e infine collegare i diversi tratti per dare origine all’intera traiettoria. Come riferimenti sono presi in considerazione le traiettorie ottime in anello aperto, generate da L.Ferrella and F.Topputo [1], su cui sono posizionati alcuni waypoints arbitrari attraverso cui la navicella deve passare grazie alla guida ZEMZEV. In seguito è impostato un problema di ottimizzazione parametrizzato rispetto alla posizione dei punti, per determinare un minimo set di waypoints necessari per avvicinarsi all’ottimalità. I risultati raggiunti sono migliori paragonati alla disposizione arbitraria dei punti quindi lo schema di guida utilizzato è capace di combinare la guida ottima in anello aperto con le capacità di targeting ZEMZEV. La stessa guida è applicata a casi più complessi – casi con un limite di spinta molto basso con conseguente controllo di traiettorie che possiedono più di 500 spirali. Infine il modello è simulato in uno scenario più realistico per verificare la robustezza del sistema utilizzando anche il metodo di Monte Carlo. La soluzione quasi-ottima dell’algoritmo ZEMZEV in questi trasferimenti orbitali (ambiente non lineare) è ampiamente verificata confrontandola con le soluzioni in anello aperto; si può quindi affermare che l’algoritmo può competere con queste ultime poiché rende il sistema più appropriato a gestire situazioni perturbate o con incertezza grazie alla retroazione in anello chiuso.
Waypoints Zemzev feedback space guidance for multi-spiral, long-duration low thrust transfers
LANAVE, GIULIA
2015/2016
Abstract
The problem of controlling the trajectory of aerospace vehicles from an arbitrary initial position and initial velocity to a desired target position with constrained, free or pointed terminal velocity in a specific time is of fundamental interest as an optimal control problem. The purpose of this thesis is to design an optimized closed loop solution in low-thrust, long duration orbit transfer environment considering two different scenarios – Earth to Mars and GTO to GEO. Low-thrust transfers are taken into account because less fuel is required when using electric thrusters which results in less launch mass and less cost. The zero-effort-miss/zero-effort-velocity feedback guidance algorithm which has great potential for autonomous onboard implementation is applied. Unfortunately it does not give an optimal solution because during the entire mission some assumptions are violated. A way to improve the performances is to divide the total flight time into more segments, determining with a rigorous optimization method near-optimal waypoints to connect the different segments. Open loop energy-fuel-optimal trajectories generated by L.Ferrella and F.Topputo [1] are considered as reference where some arbitrary points are put to be targeted by ZEMZEV guidance. An accurate study is developed valuating performances which vary with the number of waypoints. Afterwards an optimization problem, parametrized with the position of the points is set in order to determine the minimum set of waypoints necessary for close-to-fuel-optimal waypoint space guidance. Improved results of optimization are figured out and compared with the random positioning reaching the conclusion that way-point guidance scheme is able to combine open-loop optimal guidance with ZEM/ZEV targeting capabilities. The same guidance is applied also to challenging cases – i.e. cases with a very low limit thrust which results in controlling a trajectory with more than 500 spirals. Finally the model is simulated in a more realistic scenario to verify the robustness of the system performing Monte Carlo analysis. By employing the zero-effort-miss/zero-effort-velocity algorithm in the highly nonlinear orbital transfer and comparing with corresponding open-loop optimal solutions, its near-optimality is further verified. Furthermore the algorithm can even compete with corresponding open-loop optimal solutions, while its feedback characteristics make it more suitable to deal with uncertainties and perturbations.File | Dimensione | Formato | |
---|---|---|---|
Tesi_GiuliaLanave.pdf
accessibile in internet per tutti
Descrizione: Testo della tesi
Dimensione
4.85 MB
Formato
Adobe PDF
|
4.85 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/10589/133097