Speaker
Riccardo Maderna
(AIKO)
Description
The ESA-funded Health-AI project explored AI-based approaches for on-board FDIR. The Health-AI system has been developed by AIKO by leveraging the proprietary orbital_OLIVER software for onboard autonomy. It implements a combination of deep-learning and knowledge-based models for detection and classification, respectively. Use cases and data comes from real smallsat missions operated by Tyvak International. The developed system has been deployed and tested on representative flight hardware, including Ingeniars' GPU@Sat, a general-purpose GPU-like IP-core fitted on a radiation-hardened FPGA.
The presentation will discuss the project results, lessons learned, and route forward toward an operative system.
