Edge AI Acceleration for Critical Systems: From FPGA Hardware to CGRA Technology

This work discusses advancements in edge AI processing for critical systems, focusing on satellites. We explore hardware solutions for real-time and autonomous data processing using Field Programmable Gate Arrays (FPGAs) and Coarse-Grained Reconfigurable Arrays (CGRAs). The research outlines critical systems' challenges, such as power constraints, radiation tolerance, and the need for reliable AI processing. The paper details the design and implementation of an HDL-based GPU system on an FPGA and the FPGAAI framework that automates the design of AI accelerators on FPGA. Future research focuses on extending neural network support and exploring collaborations for tape-out opportunities of CGRA prototypes.