ICARUS will provide an immersive environment that allows for training of Assembly Integration and Testing (AIT) procedures in Virtual Reality (VR).

In this activity, Mission Control Space Services deployed a low-level implementation of the OPS-SAT SmartCam model using a Field Programmable Gate Array (FPGA), comparing against a high-level CPU model using Tensorflow Lite. Experiments showed that the FPGA implementation reproduced the precision and accuracy of the high-level model, while running at a slower speed. Further optimisations of the FPGA are expected to close the gap in timing and unlock new methods for deploying deep learning on spacecraft.

Along 20 years of life sciences research in space, several full size microscopes were developed (phase contrast, fluorescence analysis, and recently the Advanced Light Microscope from NASA) to be used on board the ISS. However, such bulky systems appeared to prevent a versatile use and limit the potential of experiments that can be performed today on ground. In particular, none of them can be accommodated on an on board centrifuge to perform experiments in microgravity and at various G level. And for most of them, the samples are fixed on a glass slide and allow only static analyses.

Novel GNC systems are based on on-board real-time optimized GNC systems to enable safe ascent, agile manoeuvring as well as ultraprecise stage landing on ships and ground. Modern space transportation and future exploration systems will be re-usable to contribute in reducing mission costs, while improving reliability, safety and performance. This activity shall establish novel GNC architectures and systems implementing Real-time Optimised GNC system for reliable and fault tolerant ascent flight, precise orbital guidance and control as well as for precision entry descent and landing.

In this project the authors looked at event-based sensing and processing for space situational awareness (SSA). Many advantages exist with the new paradigm of neuromorphic engineering, the authors investigate whether event-based optical data is suitable for the task and if processing with Spiking Neural Networks (SNN) provides advantages in terms of efficiency and efficacy. With the implementation of a novel heterogeneous LIF SNN, authors claim to surpass the state of the art in event-based SSA with a 15% increase in accuracy.

In applications such as space traffic and disaster response management, there are growing public and private capabilities. Multiple different stakeholders can detect and identify events (space collision predictions, natural disaster definitions). Augmentation of events definitions lead to confusion for local authorities, operators or any users of these information for costly decisions. This project, Space DAO, sets the base infrastructure to agree on consensus mechanisms to reduce confusion and augment trust in key decision making information.

With the 'democratisation of space', and the proliferation of large non-geostationary orbit constellations, the pace at which human-made objects are being deployed into orbit demands urgent action in defining and supporting a global space traffic management plan.

The LEO-GEO4GHG project studied the feasibility of new frontiers of cloud computing in space and artificial intelligence (AI) combining multiple sources of data such as the SATLANTIS GEI-SAT constellation and other external meteorological and atmospheric data with the final aim of providing near real-time methane detection and quantification.