ESA GNC Conference Papers Repository
CubeNav: A Flight Dynamics tool to support Guidance and Navigation Operations of deep-space CubeSats
CubeNav is an ASI-funded project aimed at developing a Flight Dynamics infrastructure tailored for supporting Guidance, Navigation, and Control (GNC) operations of CubeSats involved in deep-space missions. The project is led by the Radio Science and Planetary exploration Lab of the University of Bologna (Unibo) in partnership with the DART Lab of the Politecnico di Milano. CubeNav leverages on the complementary expertise of the two teams. The development of the navigation analysis tools is performed by researchers of University of Bologna while the DART Lab is in charge of the implementation of the guidance tools. Both teams have been involved in different deep-space CubeSats missions. Namely, Unibo performed the navigation of both LICIACube and ArgoMoon while the DART Lab is leading the phase B study of LUMIO and has performed the mission analysis and GNC development of HERAs Milani CubeSat. CubeNav is designed to ease the planning, development, engineering, and validation of Flight Dynamics operations and procedures, reducing the need of step-by-step intervention by satellite operators. This, in turn, could drastically reduce the learning curve to perform GNC analysis, the chances of human error, and time for scenario setting and data analysis. These aspects are identified as key elements for performing a cost-effective ground-based navigation of small, low-budget interplanetary missions such as those involving CubeSats. For those missions the system development costs scale with its size, but this is not yet the case for flight dynamics operations which are still performed from ground. Therefore, it is clear that any software that could minimize the involvement of operators in repetitive tasks will beneficially impact on navigation costs of such missions, thus potentially enabling a wider number of deep-space missions with CubeSats. This work presents the overall architecture of CubeNav, with a focus on the modules being implemented and on the level of interaction between the user and the software blocks interfaces and outputs. First, an introductory overview of the main objectives of the tool and how these have driven the definition of the overarching design criteria is given. In particular, it is described how the software is being developed in a modular fashion and how this shall ease the future implementation of an automation layer to perform operations and analyses. Then, an overview of the external dependencies and underlying third-party software libraries is provided. These include ESAs GODOT software and SPICE which are employed for the generation of orbital models and observations model. In the second part of the paper the three software functional blocks composing CubeNav are introduced and outlined: the interface and the data format conversion layer, the flight dynamics layer and the Graphical User Interface (GUI) layer. The interface layer serves as connection between Unibo and DART proprietary software tools developed in previous projects to perform navigation and trajectory optimizations. Since these tools have various output formats it is necessary to convert them in a set of standardized interface files which can be then employed within the GODOT and SPICE modules or used to exchange information with external partners. Among proprietary software which CubeNav can interface are worth mentioning DART Labs optimization tools ULTIMAT, LT2O, and DIRETTO. The interface layer also provides code segments necessary to directly convert data package obtained from these tools and make them available to the GUI for final processing and visualization. The second functional block instead groups a set of tools written in C++ and Python to perform flight dynamics computations. These are further divided into two sets: navigation and guidance. The navigation one, developed by University of Bologna, serves as an aid to the operational navigation procedure, making emphasis on the time management and real-time assessment of the orbit determination solution. The second set gathers instead tools developed by Polimis DART aiding Guidance & Control (GC) analyses. These comprises tools for the computation of astrodynamical events, which involve the celestial mechanics of bodies that the spacecraft could approach during the mission and that could influence its operativity. Furthermore, software to handle the reconstructed and predicted spacecraft state and its uncertainty in the optimization time span or to manipulate and convert the control acceleration in direction and magnitude are also part of this subset. Finally, the Graphical User Interface is introduced. The GUI is the last layer of the tool and represents a two-way interface which the user will interact with. The GUI will allow the user to specify input parameters and select desired outputs as well as providing them with the related graphical and text outputs. It is described how the way of presenting and storing these outputs is beneficial both for the interaction with the operators and for allowing the generation of a schedule of automated activities and tasks for operation purposes. Subsequently, some details are given for the timeline plot, which allows to group all relevant information coming from navigation (station passes, station availability, ...) and guidance (manoeuvre times, no-thrust windows, ) providing the users, mission analysts and FD manager with a complete overview of the upcoming events and activities. To conclude, some remarks on the ongoing development and future improvements of CubeNav are described, focusing in particular on the automation layer. Overall, CubeNav could be an effective integrated tool for supporting flight dynamics operations, efficiently exploiting the knowledge and tools of two research groups. The tool enables to obtain all mission details relevant to the navigation operations, thus raising the level of automation of flight dynamics operations. The modularity of the tool represents its strength in terms of adaptability to different optimization tools and tasks automation. The authors would like to acknowledge ASI for the support received.