Edge AI
BetaEdge
Project Inspiration
Project Inspiration
Small satellites will become more autonomous thanks to the presence of AI onboard. Applications that employ large amounts of data, like Earth Observation, have a lot of potential. The images can be analyzed onboard, in order to download just the results.
PROJECT GOALS
Reduce latency
Processing data at the edge can reduce the amount of time it takes for the results to be available, as the data does not need to be transmitted over a network to a central location.
Improve reliability
Edge AI can help to improve the reliability of Earth Observation systems by allowing them to continue operating even if the network connection is lost or degraded.
Enhance data privacy
Processing data at the edge can help to protect the privacy of the data, as it can prevent sensitive information from being transmitted over a network.
Reduce costs
Using edge AI can reduce the amount of data that needs to be transmitted, which can save on bandwidth costs and reduce the amount of infrastructure required.
ABOUT PROJECT
BetaEdge is a Space satellite payload that leverages Edge AI technology for classifying data acquired from SAR and hyperspectral cameras on Small Satellites.
A prototype has already been developed to demonstrate the feasibility and, as shown in the demo video, consists of a COTS board on which the Linux operating system was installed and a .tflite model for wildfire detection was loaded. The network was trained with an existing open source dataset. The prototype recognizes and delimits the wildfire with a calculation of its extent.
