In-Orbit Artificial Intelligence and Machine Learning On-Board Processing Solutions for Space Applications : Edge-Based and Versal Space Reference Designs : First Design-In Experiences

An increasing number of on-board processing applications require intelligent, real-time in-orbit processing to extract value-added insights, rather than clog precious RF downlinks with bandwidths of data for cloud-based, post-processing on the ground. Some applications require autonomous, immediate decision making, e.g. a LEO spacedebris retrieval spacecraft outside of its ground-station contact time would not be able to receive a late command to initiate a collision-avoidance maneuver, or a defense satellite performing space domain awareness using data from multiple sensors followed by object detection and classification may require a friend or foe decision. High-definition SAR imagery is increasingly generating huge amounts of Earth-observation data and in-orbit AI inference and the implementation of neural networks allows for feature identification and characterization to assist real-time decision-making, e.g. tracking wildfires or floods.