MOVIQ: an end-to-end approach to enable high-resolution hyperspectral vision intelligence on board satellites

Recent advancements in technology have led to the development of smaller instruments, sensors, and electronics, enabling small satellite systems to collect vast amounts of data at a relatively low cost. This has sparked significant growth in the private Earth observation (EO) industry. Despite the deployment of satellite constellations with improved spectral, spatial, and temporal resolutions, there remains a bottleneck in data transmission and downlink capabilities, hindering efficient information extraction. Emerging technologies in artificial intelligence (AI), inter-satellite communication, and edge processing are promising to address these challenges, revolutionizing satellite missions and operations. The integration of intelligent payloads in various recent satellite demonstrators is marking progress towards autonomous satellite constellations. To meet the growing demand for autonomous and intelligent remote sensing, initiatives like the MOVIQ project are crucial. MOVIQ aims to enhance data processing efficiency and advance EO capabilities through agile satellite platforms, high-resolution optical payloads, fault-tolerant AI processors, advanced applications and streamlined integration and optimization frameworks. These efforts promise to deliver cost-effective, intelligent mission platforms capable of real-time situational awareness and application-specific insights to further drive innovation in the EO sector.