High-performance on-board data processing is a strategic technology for space agencies and other space organizations. Here are a few reasons why it is crucial:
-
Data Volume Reduction: Space missions generate vast amounts of data, often exceeding the available downlink bandwidth. On-board data processing allows for real-time analysis and filtering of data, reducing the volume that needs to be transmitted to the ground. This enables more efficient use of limited communication resources and facilitates faster decision-making.
-
Latency Reduction: The distance between Earth and spacecraft can cause significant communication delays. By processing data on board, time-critical operations can be executed without waiting for commands from the ground. This is particularly important in scenarios such as autonomous navigation, hazard detection, and response to critical events.
-
Autonomy and Fault Tolerance: On-board data processing enables spacecraft to operate autonomously, making independent decisions based on processed data. This autonomy is crucial in scenarios where real-time communication with the ground is not possible or in situations where rapid responses are required. Additionally, on-board processing allows for fault detection and mitigation, enhancing the reliability and safety of space missions.
-
Science and Exploration: High-performance on-board data processing enables scientific instruments to analyze data in real-time, allowing researchers to rapidly identify interesting phenomena and optimize data collection strategies. This capability is particularly valuable for missions involving remote sensing, planetary exploration, and astronomy.
National Defense and Civil Security from Space
Through high-performance on-board data processing, the response time to critical events can be reduced, as data is available to decision makers faster. The sensitive geographic positioning of Greece creates a need for national defense applications from space. Applications of civil security are critical for the country as well, given the catastrophic phenomena that occur frequently, such as wildfires and floods. High performance on board data processing, coupled with technologies such as virtualization and flexible ground stations can provide innovative platforms that are part of a distributed network. This network can benefit remote sensing applications from space which can be implemented to answer Greece's needs from space and create a sovereign product at the same time, as the heritage already exists in the country.
Onboard Multi-Frame Super Resolution Image study
Super-resolution refers to the process of improving the spatial resolution, that is, the level of detail, of an image or an acquisition system. Our innovative idea is to use OPS-SAT's powerful on-board computer to deploy a machine learning-based super-resolution algorithm directly on-board the satellite. It has the potential of opening new perspectives of EO applications mainly in time-critical domains like safety and security. In the frame of this experiment, we successfully developed and deployed an AI-based multi-frame super resolution algorithm on board OPS-SAT achieving a notable improvement in image quality and proving the merits of the proposed approach. The study was undertook by OHB Hellas with FORTH as a subcontractor.
Cognitive Cloud Dual Camera study
The main objective of this study, undertook by OHB Hellas with FORTH as a subcontractor, is the evaluation of the merits of a dual camera setup for the acquisition of both large swath, low- resolution images and narrow-swath, high-resolution images of identified events of interest, on-board, through AI methods. The feasibility of the approach on a mission-level concept of a single-satellite, dual camera setup as well as a two-satellite (leading-trailing), single camera setup was investigated. The AI approach was demonstrated on commercially available representative high-performance on-board computers (Xilinx Kria SoM / Google Edge TPU SoM) through a state-of-the-art fire detection use case.
CAIRS 21
The goal of this project is to evaluate the suitability of COTS hardware accelerators (Google Edge TPU) for use in space missions, focusing on radiation characterisation, performance evaluation and software support allowing efficient uploading of new AI/ML models to the satellite during flight. It was developed with National Technical University of Athens as prime contractor.