Introduction
The integration of Artificial Intelligence (AI) at the edge is reshaping the aerospace and unmanned aerial vehicle (UAV) sectors, ushering in an era of unprecedented technological advancement. By enabling real-time data processing directly on devices, edge AI overcomes traditional barriers of latency and bandwidth in data transmission.
Low Power SoMs in UAVs
In the UAV industry, the adoption of low-power Systems on Modules (SoMs) marks a significant advancement. These systems provide crucial AI acceleration at the edge, enabling drones to process and analyze data in real time. This capability is transformative for various applications, including surveillance, mapping, and environmental monitoring, with the added benefit of extended operational durations due to low power consumption.
RAD-Hardened SoCs and FPGAs in Satellites
For satellites, the focus shifts to utilizing Radiation-Hardened System on Chips (SoCs) and Field-Programmable Gate Arrays (FPGAs). These components are essential in the hostile environment of space, offering resilience against radiation and extreme temperatures. This technology enables satellites to perform complex AI tasks, such as image processing and change detection, efficiently and reliably.
Innovative Applications
A practical application of this technology is evident in imaging satellites. With edge AI, these satellites can gather and process images using super-resolution techniques, selecting only the most relevant and high-quality images for transmission. This process significantly reduces the volume of data sent, optimizing both communication bandwidth and costs.
In surveillance missions, AI-enabled satellites can photograph specific ground targets on each pass. Utilizing AI algorithms, they analyze changes over time, transmitting only significant updates. This selective data transmission approach ensures efficient use of communication resources.
