Architecture of a Simulation Test Bench for Operating Large Satellite Constellations

Abstract

The approach of using FlatSats to aid the development and testing of a mission has been adopted effectively for CubeSats and small satellites. However, when considering a satellite constellation, the FlatSat approach can become limited, which is where simulating the interaction between the space and ground segments of large constellations offers a powerful enhancement. Moreover, such a simulation can be employed to evaluate new technologies and quantify their potential impact in the operation of a constellation. This paper presents an innovative architecture for a simulator designed to precisely meet these challenges. The simulator deploys the NASA Operational Simulator for Small Satellites (NOS3), building on its heritage, open source components, active development and support. The use of NOS3 as the underlying simulation engine ensures compatibility with a wide range of hardware and software systems, which facilitates the testing and validation of satellite designs. It also allows for the incorporation of additional data sources and the emulation of hardware components, thereby contributing to the realism of the simulation. Our proposed architecture addresses the challenges of simulating large satellite constellations, including scaling, computational efficiency and fidelity of the system modeling. A key feature of the simulator is its modularity, which allows for the integration of different flight software frameworks, subsystem models and ground segment software. These, in combination, permit different types of constellations to be simulated. The simulator uses a fully containerized and microservice architecture, which ensures its portability and ease of deployment across different computing environments. This also facilitates the management of dependencies, isolation and rapid scaling of the simulation environment. Furthermore, the simulator can be run on a self-hosted infrastructure, providing users with full control over their simulation environment and data. This is particularly beneficial for organizations with strict data security requirements or those wishing to integrate the simulator into their existing infrastructure. Alternatively, the architecture can be easily deployed in a cloud-based infrastructure. We also present how this architecture can be deployed as an operational simulator or as a test bench for new technologies, such as exploring the use of a Digital Twin in the operation of a satellite constellation.