Keywords: Innovative Launchers and strategies, Artificial Intelligence for on board autonomy, Artificial Intelligence for fractionated assets data management
Abstract: In a NewSpace scenario where new small missions are being implemented every day to bring concepts to space, the core interest is more and more moving towards the applications and the demonstration of technology and concepts themselves. And still access to space goes trough a number of practical issues and repeated steps that several satellite and space off-the-shelf shops and several service providers are already trying to facilitate. However, even today, with these new tools available, COTS and service selection, breadboarding, implementation and, finally, launch may easily take two years in time and considerable resources need. Here is where AIX comes in. AIX makes available satellite resources and on-board applications as-a-service. It may act as a technlogy in-orbit demonstration, but above all makes users free to have access to a satellite and to its embarked resources on demand, potentially just in days ater the need is consolidated. AIX provides the usage of the onboard components (and of a full satellite) through a set of standard and custom services. Users can command their own payload or pick-up the application they need from the AIX app store, configure and run it on the satellite. Definitevely, users need only to define what they care of, everything else is already provided. As a final result, they can focus on data and information they look for. The system will take care of scheduling the data acquisition, transforming data into actionable information and also raising near real-time alarms when services require. AIX is a gamechanger. It processes data where it’s more convenient, starting on-board at the “space edge”; it turns space products generation into services, making the satellite transparent; it makes on-board resources flexible enough to fit to different applications and address different needs, thanks to AI and DLT technologies advances. AIX fosters the transition from a traditional space model to a really commercial one, reducing bottlenecks and barriers, enabling new opportunities and enhancing the effectiveness of the services delivered to the ground. Scientists and technologists may test innovation, AI algorithms and their proof-of-concepts directly in space and prove their value.

Keywords: Earth-bounded Missions, Constellation missions
Abstract: HiVE (High-resolution VEgetation monitoring mission) is going to be the world's first microsatellite constellation for thermal infrared land surface temperature monitoring. In development by a consortium led by ConstellR GmbH, a German newspace start-up, together with OHB System, NanoAvionics and Fraunhofer EMI, its goal is to provide global land surface temperature (LST) imagery optimised for high-precision agriculture, water management, temperature-derived crop health management, yield forecasting and sustainable resource management. Sub-field crop monitoring calls for high resolution imagery, day-to-day planning requires high revisit frequencies and operational use demand low latencies for data and analytics delivery. Due to its commercial nature, the constellation aims to be cost-efficient via the use of commercial-off-the-shelves components, while providing all key capabilities. The HiVE constellation mission architecture comprises multiple operational concepts, including virtual calibration for payload miniaturization and dynamic (on-demand) tasking/targeting, serving more users/payloads. By introducing novel operational concepts, the required payload mass and volume can be minimised, thereby reducing space segment costs to a fraction of current systems whilst ensuring high radiometric precision. In the paper, the system architecture and constellation-level trade-offs are described with particular focus on how to meet customer requirements via satellite design, orbital configuration and operational concept. The constraints and system-level impacts coming from both COTS use and rideshares launch opportunities are discussed.

Keywords: Constellation missions, Non-Earth Missions, Constellation traffic management
Abstract: Over the last decade, the scientific interest in Mars drastically increased. The planned growth of the number of robotic missions, together with the sensors' increasing data acquisition capabilities and the expected crewed expeditions, entails a significant increase in data flow between the martian assets and Earth both in volume and frequency of contact. In particular, crewed missions would lead to the need for nearly continuous communication with Martian assets. The keystone to avoid the future martian telecommunication deadlock stays in specialising assets on specific functionalities through infrastructures: the paper proposes a distributed Mars-based orbiting system servicing as a communication relay for any scientific and technological mission operating in the red planet environment. The paper explores the design of a small satellites Martian constellation to maximise the surface coverage and visibility time with respect to ground users while reducing the station keeping efforts of the assets. The paper also discusses different strategies to reach the Martian environment and deploy the constellation considering both standalone and shared launches. The collection of transfer and operative orbits options were traded off in terms of cost-effectiveness to detect the suggested baseline for the constellation, which is presented with its performances. Once the architecture is selected according to coverage and station keeping constraints, the paper shows the results of a proposed self-standing state reconstruction approach for the constellation assets. Indeed, the peculiar dynamics induced by the Mars-Phobos and Mars-Deimos multi-body regimes lead to the possibility to exploit the Linked, Autonomous, Interplanetary Satellite Orbit Navigation (LiAISON) method: only relative range and range-rate measurements are enough to reconstruct the satellites state vector.

Keywords: Low-thrust orbit acquisition, Fomation Orbital Injection and Acquisition, Earth-bounded Missions
Abstract: The framework of this paper is a CNES demonstration mission called “ULID” which is dedicated to the moisture and ocean salinity monitoring using L-band interferometry. This mission consists of 3 nanosatellites flying in formation. The distance between the satellites in operational configuration is about 40 meters and that’s one of the challenging aspects of the mission. The paper describes how the geometry of the formation has been designed. It details the acquisition as well as the maintenance strategies.

Keywords: Non-Earth Missions, Perturbations Analysis, Low-thrust orbit acquisition
Abstract: Over the last decades, the exploration of Phobos and Deimos has acquired relevance from both scientific and human missions standpoint. Both moons can play a key role on supporting the Martian Exploration program improving teleoperation capabilities and infrastructure support. The purpose of this work is to propose a different approach to the problem of global longitudinal coverage of Mars, exploiting the characteristic of natural satellites of the planet itself. The proposed constellation can be an alternative to the Areostationary Martian Orbit (MAO). In fact, three satellites equally spaced at Deimos height are enough to provide global longitudinal coverage, with a reduced cost. The main source of perturbation at areostationary height is due to Mars J22 harmonic, which requires high maintenance cost for an equivalent configuration in MAO. By leveraging Lyapunov orbits in the Mars-Deimos perturbed Three Body Problem, it's possible to achieve long term stable orbits with lower cost, as well as ballistic orbits with bounded variations. The characteristics and the performance of the constellation are analyzed, as well as the transfer orbits between the Mars-Deimos Lagrange points leveraging resonant orbits. The possibility of placing an extra satellite in an orbit around Phobos with the purpose of increasing performance is also detailed, as well as the transfer trajectory from the Deimos L3 point to a DRO orbit around Phobos.

Keywords: Inertial Navigation, Sensor fusion, Modelling and Parametrization of Relative Dynamics
Abstract: Relative baseline estimation is an integral part of satellite formation flying missions. GNSS-based relative positioning has been a dominating technology for formation missions in LEO, where very precise estimates could be obtained for formations with small inter-satellite distances (1-10 km). Larger baselines between the satellites (larger than 10 km) pose the problem of huge differences in the ionospheric delays experienced by the signals received by each receiver. This problem could be mitigated by using precise ionospheric-free combinations that could only be obtained by dual-frequency receivers, which is not a cost efficient option for the modern low-cost miniature missions. In this paper, the problem of relative baseline vector estimation is addressed for formation missions with large inter-satellite distances occupied with single-frequency receivers. The problem is approached using the space-proven relatively simple Extended Kalman Filter with an advantageous setting for the observation vector.

Keywords: Modelling and Parametrization of Relative Dynamics, Formation Flying missions, Optimal Control
Abstract: In the next future, many missions will target the Moon as a future step for space exploration. The first leap to reach this new objective is to send a crewed orbiting space station on an L2 Near Rectilinear Halo Orbit around the Moon. The station will need multiple resupplying missions to guarantee the life on board, therefore the design of efficient trajectories to safely phase rendezvous and dock with the space station becomes a key point in the mission design process. Therefore the work proposes the design of a phasing strategy to approach the passive target space station, the strategy is based on the use of butterfly orbits as parking orbit, with the intent to recover some phase and on the design of an efficient impulse sequence to transfer from the parking orbit to the target one. The work investigates three different orbits in the butterfly family as parking orbits and selects the best in terms of fuel consumption and the total time of flight. The paper also proposes two different guidances algorithms to transfer from orbit to orbit: cr3bp-Lambert and optimal. All the results are evaluated on the basis of clearly stated criteria and the best one is extensively detailed.

Keywords: Optimal Control, Modelling and Parametrization of Relative Dynamics, Innovative solutions for Intelligent Scheduling and Operations
Abstract: Very Low Earth Orbits (VLEO) are currently object of renewed interest, because of their low cost for orbit injection, potential higher performance for remote sensing, and possibility to use small-class launchers. These orbits are deeply affected by aerodynamics forces, that must be properly modeled both in their drag and lift component. The orbital decay for low altitudes is very fast, and a minimum altitude for performing missions of reasonable duration is first assessed. The performance that can be obtained by a single satellite in VLEO can be further and greatly enhanced if a companion satellite is released on the same orbit, thus realizing a formation of two (or more) micro-satellites. In this paper, the use of mobile appendages for producing differential forces is considered. Thus, a double layer control is developed, in which a traditional “common mode” active control is used for orbital keeping, while a “formation mode” control using aerodynamics forces is developed for formation keeping and reconfiguration. A model predictive control is developed for scheduling the actions for formation control. The prediction of the future relative state of course requires an accurate but simple dynamics model, so that the optimization can run fast and in principle even in on-board autonomous applications. The selected linear model uses differential orbital elements instead of usual cartesian coordinates because it allows for an easy and reliable inclusion of the main perturbation (differential J2 and drag). The optimization algorithm used for computing the control schedule, which is realized by means of alternate rotations of the mobile panels, is based on linear programming techniques. The performance of the proposed control in terms of accuracy and cost is shown and discussed together with the limitations that are intrinsic to this method.

Keywords: Optimal Control, Fomation Orbital Injection and Acquisition, Distributed system deployment strategies
Abstract: The problem is considered to find the time-optimal control that transfers a 4th-order system that consists of a double integrator and a harmonic oscillator, coupled by one control, between two arbitrary states. That 4th-order system is equivalent to the Hill-Clohessy-Wiltshire relative dynamics of an orbiting spacecraft in proximity of a second spacecraft on a circular orbit, subjected to a thrust parallel to the orbital velocity. A new method is introduced to determine the time-optimal control problem stated above. This method combines two previously discovered optimal control synthesis methods: the method by Romano-Curti, that enables to find the optimal control transferring a general Linear Time Invariant Normal system between two arbitrary states; and, the method by Belousova-Zarkh, that enables to find the optimal control transferring a 4th-order system consisting of a double integrator and a harmonic oscillator, coupled by one control channel, from an arbitrary initial state to the origin of the state space, if the optimal control is a priori known, that transfers the same system from a reference state to the origin. The here proposed combined method utilizes two phases. First, a number of reference minimum-time controls are obtained that transfer the system from specific reference states to the state space origin; this is achieved by exploiting Pontryagin’s principle and back-propagation from the origin. Second, a search is run along a curve in the state space that depends on the desired boundary states. A minimum-time control problem is iteratively solved to find the optimal control history that steers the system from a state on that curve to the origin, by exploiting Belousova-Zarkh method, until a particular state is found which satisfies an equivalency condition that, as demonstrated by Romano-Curti, guarantees that the optimal control history pertaining to the problem of transferring the system from that state to the origin is the same optimal control history that transfers the system between the desired boundary states. The new results, tested by numerical experiments, have both a theoretical and a practical value, as they could be applied to the optimal guidance of autonomous spacecraft.

Keywords: Nonlinear filtering
Abstract: Abstract: The long-term and high-precision formation navigation problem of distributed satellites is addressed here through an adaptive multi-mode fusion formation navigation algorithm. Firstly, an extended Kalman filter is created using the relative kinematics equation and high-precision differential GPS measurement data to obtain a set of formation configuration parameters. Secondly, an adaptive extended Kalman filter based on the relative orbital elements is established to obtain another group of formation configuration parameters, taking into account the relative motion model's error and the instability of the real-time measurement data. Finally, multi-mode fusion is used to address the issue of relative navigation instability and invalid long-term measurement data. The simulation results show that the method is capable of overcoming the model linearization error, achieving higher configuration determination accuracy than the traditional navigation algorithm, and achieving long-term stable formation navigation. keywords: formation navigation, adaptive multi-mode fusion, Kalman filter, relative orbital elements, 1 Introduction With the development of satellite technology, there are more and more distributed formation satellites with collaborative work, and inSAR satellite systems are one of them. In the InSAR formation mission, through the formation configuration, various functions that are difficult for a single satellite to complete can be realized, including high-resolution SAR imaging and ground maneuver target indication. In order to better realize these functions, the determination of the formation configuration has become the focus of research. At present, the determination of the formation configuration is mainly achieved by relative navigation algorithms, the traditional method is mainly based on the CW equation or Hill equation, but the method has a variety of assumptions, not to meet the assumptions there is a large error, at the same time, in practical applications, real-time measurement data instability, data for a long time invalidity and other aspects are easy to cause filter divergence or error, and even cause the formation control system collision avoidance, resulting in serious consequence

Keywords: Mega constellations, Modelling and Parametrization of Relative Dynamics, Collaborative distributed decision making
Abstract: Mega constellations have larger number of satellites and higher distribution density compared with traditional constellations, that leads to challenges in maintenance control. This paper weakens the concept of constellation configuration, and innovatively presents the self-organizing maintenance control method. Satellites are controlled based on the local information of adjacent satellites rather than the exact configuration. The continuous coverage constraint is reformulated to constraints of right ascension of ascending node and relative motion bound between every two adjacent coplanar satellites. The constellation is maintained by coordinating the semi-major axes of coplanar adjacent satellites. The relationship between continuous coverage constraint and semi-major axis difference is studied. The semi-major axis adjustment strategy for a single satellite is proposed, and the autonomous semi-major axis control rules for coplanar satellites are determined. The proposed maintenance control method is verified in GW-2 constellation. Results indicate its effectiveness and good performance in control frequency and consumption.

Keywords: Mega constellations, Constellation missions
Abstract: With the advancement of communications technology, lower launch costs, and development of small satellites, the satellite internet marketplace is becoming a complex and fast-expanding sector. While modeling of individual satellite constellations and comparisons of relative technical performance have been conducted, modeling the competition between satellite internet providers and their complex strategies remains largely unexplored. This project aims to model the dynamics of long-term market strategies and competition between different satellite internet providers by framing the problem as a multi-player, strategy-simulation game. We use a mixture of modeling and gamification to create the prototype of a game in which players construct and operate proliferated low-Earth orbit (P-LEO) constellations that provide satellite internet to simulated customers. In the gamified environment, players can take four basic actions: ordering satellites, building ground stations, launching satellites into orbit, and setting a monthly internet subscription price for their customers. The framework also includes a primitive technology tree where players can explore the impact of new technologies for better constellation performance and customer acquisition as emergent game strategies. Like previous strategy simulation games designed for research, a server authoritative software architecture is used to connect a hosted web server to multiple clients. The envisioned outcome of this project can be used as an educational tool, a business decision support system, a simulation environment for national security studies, an AI test environment, or simply a recreational game.

Keywords: Mission Analysis tool for Mission Control, Mega constellations, Distributed system deployment strategies
Abstract: Reconfigurable satellite constellations represent a great opportunity to overcome limitations coming from the current mission design philosophy, offering enhanced flexibility and enabling to focus the available resources on changing mission objectives. In this work, a modular approach is envisioned to deal with the complexity of a reconfiguration strategy, tackling separately the design of the optimal constellation geometry to satisfy given observation requirements and the computation of the reconfiguration maneuvers. This paper investigates the design and optimization of asymmetric constellations in LEO, proposing a design strategy able to maximize observation performances over a user-defined area. A multi-objective optimization is chosen to deal with the complexity of the design problem and a Genetic Algorithm is employed to compute the best achievable configuration with respect to the mission objectives. The Genetic Algorithm guarantees an efficient design of the constellation, in which metrics of different nature can be combined in complex solution spaces that are not limited by analytical representations. Efforts are made to speed up the execution of the algorithm by relying on a multi-processing parallel implementation.

Keywords: Innovative technologies for distributed systems, Optimal Control, Modelling and Parametrization of Relative Dynamics
Abstract: State-of-the-art modalities of scientific observations will be unlocked through the advancement of essential technologies required to enable a formation of spacecraft to safely orbit a planetary body. The Distributed Element Beamformer Radar for Ice and Subsurface sounding (DEBRIS) aims to implement a Synthetic Aperture Radar (SAR) in a formation in Low Earth Orbit (LEO), to improve the radar’s spatial resolution and sounding investigation depth. Low-power electronics and low-cost access to space are unlocking new architectures for radar remote sensing. Implementing a radar array using distributed spacecraft requires precision co-pointing, and collision avoidance strategies for dense formations of paramount importance. Enabling these architectures through multiple Cubesats in LEO, or for interplanetary missions, requires a radar pointing capability with a high-degree of precision to ensure coherency of the radar signal across the array. Stable relative positions of the spacecraft are critical to the performance of the radar array in formation flight. More importantly, for mission safety, the spacecraft autonomy must guarantee a minimum relative distance to all other members of the array throughout the mission lifetime. These constraints raise the need for advanced Attitude Determination and Control System (ADCS) and Guidance Navigation and Control (GNC) design, capable of providing the required pointing accuracy while meeting mission specifications. An optimization based collision avoidance approach is taken in this paper to meet implementation cost constraints. The performance evaluations consider commercially available state-of-the-art systems. The spacecrafts’ positions and their orientation knowledge for each elements of the array need to be communicated efficiently between the array members through appropriate inter-array timing and synchronization techniques. The development of these technologies is essential in enabling the implementation of a sparse radar aperture for space borne exploration. In this paper we show that, using current ADCS capabilities, the required performance to enable distributed radar observations in LEO can be attained.

Keywords: Earth-bounded Missions, Mega constellations
Abstract: The Earth Observation (EO) data and services market should reach USD 8.5 billion by 2026 based on current growth trajectories. This opens to a rich scenario where EO satellites are the main characters of an industry that assures revenues and returns. These elements attract public and private investors. The number of images and data collected by EO satellites will reach in the next decade 486 PB approximatively due to the digital transformation which makes spectral, spatial, and radiometric resolutions more accurate. This poses a serious challenge to downlink data transmission capability. To solve the issue, the IntraSat project proposes to supply as the first provider in Europe the service of near real-time communication between ground segment and Low Earth Orbit (LEO) satellites through a LEO constellation of relay satellites. The project aims to provide a complete study of the overall concept, from the market and possible customer analysis to the requirements definition, including the constellation design, and technologies selection to model the single satellite components. After obtaining the real pain and gain of possible customers by Value Proposition Canvas, the paper shows the Business Model Canvas (BMC) evaluating costs and revenues. The revenue streams have been identified as the result of downlink service in different subscriptions and API services. To optimize the design of IntraSat constellation, it has been implemented a genetic algorithm (GA) tested on different fitness functions according to Analytical Hierarchy Process (AHP), used to detect experts’ opinions. The metrics play a pivotal role, guiding the solutions and reflecting the trade-off optimizations to be competitive against ground stations as a service. In the end, a preliminary design definiton has been outlined following the ESA Generic Product Tree (GPT) scheme for the constitution of satellite subsystems and to highlight the Technology Readiness Level (TRL) of the main technology expected to form the constellation satellites. The entire paper details the feasibility of the project, evidencing its strengths, evaluating the results of GA customized for IntraSat, and identifying the technology for satellite building.