A3
Collaboration with Academic Institutions and Research Centres
DEIB Role: Coordinator
Start date: 2025-04-15
Length: 24 months
Project abstract
The objective of the A3 project is to develop and validate, within a representative environment, a guidance, navigation, and control module for platforms capable of performing on-orbit servicing operations in a fully autonomous manner. The system will be based on hybrid artificial intelligence techniques to manage different levels of reasoning within the onboard system. Furthermore, the project aims to demonstrate, through representative hardware, a functional verification aligned with the desired Technology Readiness Level (TRL) 3/4.
The key goal is to achieve rapid reconfigurability of relative trajectories, essential for rendezvous operations, as well as robustness and adaptability during the close-proximity phase, and the control of assembled structures, ensuring risk minimization in highly variable operational scenarios (e.g., configuration, mass, dynamics). Starting from simple chaser-target scenarios, the project aims to progress toward complex cases involving the coordination of multiple chasers orbiting in formation, for the assembly and servicing of large structures (e.g., reflectors, large-aperture mirrors).
The study will employ soft computing and Artificial Intelligence techniques to enhance the satellite’s environmental interpretation capabilities, implementing a “spin-in” of technologies already used on Earth, thus enabling increased onboard autonomy.
To validate these innovative technologies, a comprehensive testing, characterization, and verification campaign will be conducted, applying the techniques in space-representative scenarios. This study intends to initiate the process by developing and testing in the laboratory to reach TRL 3–4.
During the project, a test platform will be developed, based on RISC-V architecture paired with FPGA-based accelerators, as a tangible demonstrator of the possibility to achieve significant performance even using commercial off-the-shelf components. The project is developed and coordinated by HEAPLab at the Department of Electronics, Information and Bioengineering - Politecnico di Milano in collaboration with the Department of Aerospace Science and Technology and Leonardo SpA.
The key goal is to achieve rapid reconfigurability of relative trajectories, essential for rendezvous operations, as well as robustness and adaptability during the close-proximity phase, and the control of assembled structures, ensuring risk minimization in highly variable operational scenarios (e.g., configuration, mass, dynamics). Starting from simple chaser-target scenarios, the project aims to progress toward complex cases involving the coordination of multiple chasers orbiting in formation, for the assembly and servicing of large structures (e.g., reflectors, large-aperture mirrors).
The study will employ soft computing and Artificial Intelligence techniques to enhance the satellite’s environmental interpretation capabilities, implementing a “spin-in” of technologies already used on Earth, thus enabling increased onboard autonomy.
To validate these innovative technologies, a comprehensive testing, characterization, and verification campaign will be conducted, applying the techniques in space-representative scenarios. This study intends to initiate the process by developing and testing in the laboratory to reach TRL 3–4.
During the project, a test platform will be developed, based on RISC-V architecture paired with FPGA-based accelerators, as a tangible demonstrator of the possibility to achieve significant performance even using commercial off-the-shelf components. The project is developed and coordinated by HEAPLab at the Department of Electronics, Information and Bioengineering - Politecnico di Milano in collaboration with the Department of Aerospace Science and Technology and Leonardo SpA.
