Scientific Work Packages

---

Work Package Two has the goal of implementing and analysing ML algorithms. 

The applications considered will allow: the autonomous triggering of special measurement modes and the selective downlink of plasma environment parameters; the advanced on-board data analysis of three-dimensional particle distribution functions; the on-board analysis of solar images; and the on-board prediction capability of SEP related hazards. 

Work Package Three focuses on developing an accelerator capable of executing core computational patterns in the AI/ML methods introduced in work-package two. 

The work package relies on traditional methodology in Electronic Design Automation (EDA), including the use of High-Level Synthesis (HLS) tools for crafting to overall architecture, the use of Hardware Description Languages (HDLs, VHDL) to design custom data-paths for parts of the accelerator pipeline, as well as synthesis and place-and-route tools to evaluate area and performance characteristics of the design.

Work Package Four focuses on developing of a dedicated software framework for the porting of existent ML models directly on the prototype FPGA designs developed in WP3 and on defining and executing the validation of the prototype through functional and performance tests. 

As for the first activity, a virtual Interface Layer will be developed which connects the high-level framework and the HW developed in WP 3.

Work Package Five will provide support to the identification and the implementation of the Machine Learning algorithms for the automation of operations on board space missions, through a validation process based on the combination of synthetic measurements and past real mission data.

Work Package One provides the overarching management function for the ASAP project. 

Work Package Six handles the dissemination activities for the ASAP Project. This includes the website, social media presence (LinkedIn), organising workshops and summer schools.