Klepsydra takes on a € 175K contract from the European Space Agency

Satellite onboard processing is experiencing significant growth across diverse sectors, including telecom, Earth observation, navigation, exploration, and science. While current onboard processing techniques mainly rely on conventional methods such as filtering, FFT, encoding, and modulation, there is a rising interest in harnessing machine learning and neural network approaches to tackle complex challenges like pattern recognition and image segmentation. Moreover, the expansion of edge processing in, for instance, satellites is predicted to see rapid development in the coming years.

HPDP-40, initiated by the European Space Agency (ESA) and DLR, is a radiation-hardened, high-performance co-processor designed to address the need for a flexible and re-programmable high-performance data processor in space applications. The solution has low power consumption, high flexibility, on-the-fly re-programmability, compact form, and low mass. It opens up new possibilities for advanced data processing in space. Tests have been conducted on Artificial Intelligence (AI) using the HPDP, albeit with a limited scope due to the lack of a generic AI framework tailored for this co-processor. 

ESA has now contracted Swiss startup Klepsydra Technologies to develop high-performance radiation-hardened onboard AI processing capabilities. The startup develops software products, which, when deployed on widely used processors and operating systems, result in a substantial tenfold increase in data processing rate and a 50% reduction in energy consumption.

At the core of Klepsydra’s offerings lies the Streaming Distribution Optimiser (SDO), an integral part of Klepsydra AI. The SDO employs Genetic Algorithms (GA) to identify the optimal distribution of computing resources for executing AI models on the target processor, considering critical factors such as latency, power consumption, and data rate.

Besides significantly simplifying AI model deployment on the HPDP, Klepsydra’s solution ensures compatibility with major AI frameworks, eliminating the need for HPDP-specific coding. This approach will enable further AI applications in future ESA missions that demand high availability, including landers and deep space robotics. With their new partnership and project, Klepsysdra and ESA could establish a European hardware/software stack, reinforcing Europe’s capability to leverage AI in space.

(Press release/RAN)