Energy-aware comparative analysis of CNN architectures for space mission inference with an evaluation of governance and sustainability

Tasdelen, Abdulkadir and Duran, Talih and Doğrul, Mürsel and Habaebi, Mohamed Hadi (2026) Energy-aware comparative analysis of CNN architectures for space mission inference with an evaluation of governance and sustainability. In: 10th International Conference on Mechatronics Engineering (ICOM'26), 8-9 September 2026, KOE, IIUM.

PDF - Published Version Restricted to Repository staff only Download (1MB) | Request a copy

Abstract

Does integrating artificial intelligence (AI) into space missions require increasingly energy-efficient onboard processing under strict power and thermal constraints in longduration autonomous operational environments? This study evaluates the energy consumption during the inference phase of nine convolutional neural network (CNN) architectures evaluated under CPU-only execution conditions in a controlled and repeatable experimental environment. Energy consumption was empirically measured using the CodeCarbon framework and normalized to per-inference energy values based on execution time and workload volume. The results show substantial empirical variability between the architectures, with energy consumption differing by up to sixfold between lightweight and computationally intensive models. Annualized, mission-scale projections further demonstrate how these differences in energy usage accumulate to have a significant impact on operations. These results show that choosing a model architecture is not just a software optimization decision, but also a mission-level design parameter that influences energy budgeting, system sustainability and long-term operational planning for AI-enabled space systems.

Item Type:	Proceeding Paper (Plenary Papers)
Uncontrolled Keywords:	Energy-aware AI, Space Missions, CNN Architectures, Green AI, Onboard Inference, Energy Efficiency, Sustainable Space Systems
Subjects:	T Technology > TK Electrical engineering. Electronics Nuclear engineering > TK7800 Electronics. Computer engineering. Computer hardware. Photoelectronic devices > TK7885 Computer engineering
Kulliyyahs/Centres/Divisions/Institutes (Can select more than one option. Press CONTROL button):	Kulliyyah of Engineering > Department of Electrical and Computer Engineering Kulliyyah of Engineering
Depositing User:	Dr. Mohamed Hadi Habaebi
Date Deposited:	05 Nov 2026 16:09
Last Update:	05 Nov 2026 16:09
Queue Number:	2026-10-Q6055
URI:	http://irep.iium.edu.my/id/eprint/132506
Indexed In:	Google Scholar

Actions (login required)

View Item