Background and justification:

.

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.

The ESA Deep Space Network consists of three 35m beam waveguide antennas located around the globe. The 35m beam waveguide antennas employ an X-Band feed covering the Deep Space uplink band near 7.2 GHz and down-link band near 8.45 GHz. Simultaneous uplink commanding at 20 kW and low noise reception (Generation #2) is supported. In order to increase the transmit power capability of ESA Deep Space antennas it is currently considered to add a new high power feed and associated 80 kW High Power Amplifier.

The driver behind the MoBaTe activity was to propose ways to improve the current software development and validation lifecycle using MBSE techniques to shift some of the system validation activities to the early stages of development. The roadmap to achieve this included identifying strategies, required model properties, modelling tools and languages needed to move from a requirements-based approach to a model-based approach. All this considering existing and established workflows and tools currently used by ESA.

As new software tools are developed for Model-Based Systems Engineering (MBSE) applications, parallel improvements are needed for Concurrent Engineering teams, since this can be one of the first steps for a model-based approach to build-up on throughout the continuation of the project. One of the main constraints during Concurrent Design (CD) studies is the capability of generating and evaluating a large amount of possible mission architectures without the implication of additional experts or time allocation.

The results of the IOSPR maturation phase study done by D-Orbit (UK) are presented.

ESA's OPS-SAT mission is a key element for operational validation in flight (or IOD/IOV) of the DeepCube service of Deep Learning at the edge that Agenium Space is preparing with the support of the GSTP programme (GSTP Make, ESA-CNES) and the R&D performed in the CORTEX project (Permanent Open Call EOEP-4, ESA/Phi-Lab, https://esacortexproject.agenium-space.com). The goal of this activity was to execute on board the inference of the simplified models defined in those projects.