This content is not published

A Comparative Study Of High-Level And Low-Level Implementations Of Deep Learning Models For Spacecraft

Thu, 06/22/2023 - 11:39
Start Date: 
2022
Programme: 
Discovery
End Date: 
2022
Programme Reference: 
21-D-S-OPS-02-e
Country: 
Canada
Contractor: 
Mission Control Space Services
Status: 
Closed
Description: 

In this activity, Mission Control Space Services deployed a low-level implementation of the OPS-SAT SmartCam model using a Field Programmable Gate Array (FPGA), comparing against a high-level CPU model using Tensorflow Lite. Experiments showed that the FPGA implementation reproduced the precision and accuracy of the high-level model, while running at a slower speed. Further optimisations of the FPGA are expected to close the gap in timing and unlock new methods for deploying deep learning on spacecraft.

Application Domain: 
Generic Technologies
Technology Domain: 
1 - On-board Data Subsystems
9 - Mission Operation and Ground Data Systems
Competence Domain: 
3-Avionic Systems
8-Ground Systems and Mission Operations
Keywords: 
OPS-SAT
FPGA
SmartCam model
Executive summary: 
Final presentation: