In this tutorial, we'll delve into multi-agent coordination techniques to manage robotic systems like satellites, rovers, and drones. We'll start by laying the theoretical groundwork, focusing on cooperative auctions and distributed optimization. Then, we'll explore practical applications, including coordinating Earth observation satellite constellations to fulfill complex, repetitive tasks. Finally, we'll demonstrate how these techniques can be adapted to ground-air exploration scenarios.

Space ROS is an open-source, community-driven, platform-independent framework derived from ROS 2 tailored to meet the demands of safety-critical space robotics applications.
It extends the capabilities of traditional ROS and enables the reuse of ROS 2 applications with minimal modifications, offering enhanced reliability, fault tolerance, and real-time performance.
This seminar will provide an introduction to Space ROS, giving an overview of its architecture, key features, and advantages, followed by a hands-on tutorial demonstrating its practical applications and showcasing the Space ROS in action on one of the simulated space environments provided by this framework.

In this session, we will provide an introduction to the requirements engineering process, focusing on elicitation and formalisation using NASA's Formal Requirements Elicitation Tool (FRET). FRET is an open source tool, developed at NASA Ames, for writing, understanding, formalizing, and analyzing requirements. In practice, requirements are typically written in natural language, which is ambiguous and consequently not amenable to formal analysis. Since formal, mathematical notations are unintuitive, requirements in FRET are entered in a restricted, natural language, called FRETish with precise unambiguous meaning. FRET helps users write FRETish requirements both by providing grammar information and examples during editing, but also through English and diagrammatic explanations to clarify subtle semantic issues. For each requirement, FRET automatically produces formalizations and supports interactive simulation of produced formalizations to ensure that they capture user intentions. We will use a number of case studies from aerospace and robotics domains to guide our discussion.

In this session, we will show how to use FRET's realizability checker to identify issues in requirement sets. In particular, FRET supports realizability analysis of requirements for identifying conflicting requirements. A realizable set of requirements describes a system that is implementable; i.e., for all expected environment behavior (e.g., input), there is a system behavior (e.g., output) that satisfies all of the requirements. Essentially, checking that a requirement set is realizable we make sure that there are no conflicts between the requirements in the set. Realizability is a stronger notion of conflict analysis than consistency checking.

FRET also connects to analysis tools by exporting verification code. We will also demonstrate how we can use FRET's support for CoCoSim to generate verification conditions from requirements that can be used to model check Simulink Models.

In this session, we will provide some exercises for the attendees to carry out focusing on aerospace and robotics applications.

In this session, we will describe our current and ongoing extensions of FRET, including how we incorporate probabilities into requirements and refactoring of requirements.

This presentation aims to exaplain advancements in the theory and application of Guidance, Navigation, and Control (GNC) algorithms, as well as mission management methodologies, and how they contribute to enhancing the autonomy of robotic systems, particularly satellites.
The development of autonomous satellite capabilities holds the potential to revolutionize mission concepts, enabling unprecedented operational flexibility and functionality. The talk focuses on the role of GNC and planning in enabling multi-agent systems to perform complex proximity operations in the challenging dynamical conditions, such as the one, near the Moon, characterized by highly non-Keplerian perturbed dynamics.

This talk will be divided into two parts: in the first one, we will introduce robust model predictive control (MPC) algorithms for kinematic and dynamic trajectory tracking, so-called Corridor MPC, followed by centralized and decentralized collaborative load transportation schemes for microgravity. In the second part, we will discuss open-source space robotics platforms with a brief literature review of the available facilities, as well as introduce the KTH Space Robotics Laboratory along with its open-source software and hardware components.

The space sector has been experiencing a real (industrial) revolution in recent years. New digital technologies, together with the large availability of capital that large internet players have forcefully deployed in the sector, are modifying products and production processes. The ambition, no longer hidden, is to change the way in which Space has been thought of up to now: from a Space that offers services defined by a few commercial players, if not institutional, to a Space that creates commercial services itself and where these services are available to many types of users. In this perspective, just as cyberspace transformed the world at the beginning of the third millennium, space is destined to change the future. The founding element of this revolution is indisputably Innovation.
Today we are already at Space 4.0 where new space applications and the potential services provided are starting to reveal previously unknown opportunities with a significant impact on the economy and on our everyday life (just think of satellite navigation) stimulating global growth.
This is mainly due to the increasingly widespread use that is being made of digital technologies that enable and generate on the one hand new applications and therefore services and on the other they overturn, in a positive sense obviously, the current production systems: it is the time of additive manufacturing, of augmented and virtual reality, of connected objects and of the introduction of robots and cobots that work in collaboration with human beings, it is the time, finally, of Artificial Intelligence that can offer significant opportunities such as a higher level of product quality at lower costs and in shorter times. And all this in a sector of excellence that for decades has followed its somewhat rigid dynamics influenced above all by the spending capacity of the Institutions.
Thales Alenia Space has long been at the forefront, as a promoter in the adoption of the main technological evolutions (revolutions) in the space field, which we have already mentioned: additive manufacturing, virtual and augmented reality, cloud, big data, Artificial Intelligence, robotics, cobotics, TWIN-X, SOLAR, COLROBOT and digital twins: today they are all part of a global strategic approach, which uses innovation and digital technology to shape the factory of the future.
Reflecting on our business case, on Thales Alenia Space, on the digital transformation of the company, this Factory 4.0 approach was designed and implemented years ago to make the latest technologies an integral part of our production processes.
At Thales Alenia Space, innovation is an integral part of our DNA and increasingly at the heart of our production process. We benefit from a process of continuous improvement of innovation, started over 40 years ago, which has allowed the company and, transversally, all generations of employees to exploit the diversity of technological firsts in Space. That said, in the recovery phase of our country, it was necessary to decide whether to invest and ride the wave of innovation or to be overwhelmed by new transitions. We have proudly designed and launched the Space Smart Factory. the concept of Innovation in the way of producing and managing the entire life cycle of a product in an open factory (the so-called Factory as a service) is now a reality: the Space Smart Factory, whose construction began in 2023, will be completed in mid-2025 and will start assembling satellites.
One of the largest digital and reconfigurable facilities of its kind in Europe. The facility will be part of a system of interconnected space factories in Italy, which use advanced technologies to build satellites of different sizes for various fields and applications. Over 100 million euros are invested in the Space Smart Factory, including funding from the Italian Space Agency (ASI) through the PNRR.
Equipped with cutting-edge digital technologies, the Space Smart Factory can be reconfigured to meet different production needs. It will be equipped with highly versatile and flexible clean rooms to support the integration and testing of a wide range of satellites of different types and purposes such as Earth observation, navigation and communications. It will be a true digital hub that will use advanced technologies in all phases of satellite construction, namely design, assembly, integration and testing, including numerical modeling and Digital Twin, Virtual and Augmented Reality technologies and simulators integrated with the supply chain and automation (robots and cobots).
The Space Smart Factory includes a Space Joint Lab which is a completely dynamic collaborative environment designed to host a variety of functions aimed at training new professional figures in the field of space disciplines and developing innovative ideas and products. It is a collaborative ecosystem that aims to be a point of connection between Research and Development institutions, Universities, Start-Ups, Suppliers, SMEs and other Industrial Partners from all over the country; even those outside the current Space supply chain.

Bridging the gap between requirement formulation and systematic testing is crucial for the specification and validation of robotic applications. This process often involves manual, error-prone tasks that become increasingly complex as the requirements, design, and implementation evolve. To address this challenge systematically, I suggest expanding behavior-driven development (BDD) to define and verify acceptance criteria for robotic systems. I present how domain-specific modeling enables us to express robotic-specific concepts and how acceptance models represented as knowledge graphs enable robust querying and manipulation, which facilitates the generation of executable testing models. I will demonstrate how our tool has the potential for automated generation and execution of acceptance tests through a software architecture that combines a BDD framework, Isaac Sim, and model transformations, with a focus on the acceptance criteria for advanced robot manipulation applications.

Bridging the gap between requirement formulation and systematic testing is crucial for the specification and validation of robotic applications. This process often involves manual, error-prone tasks that become increasingly complex as the requirements, design, and implementation evolve. To address this challenge systematically, I suggest expanding behavior-driven development (BDD) to define and verify acceptance criteria for robotic systems. I present how domain-specific modeling enables us to express robotic-specific concepts and how acceptance models represented as knowledge graphs enable robust querying and manipulation, which facilitates the generation of executable testing models. I will demonstrate how our tool has the potential for automated generation and execution of acceptance tests through a software architecture that combines a BDD framework, Isaac Sim, and model transformations, with a focus on the acceptance criteria for advanced robot manipulation applications.