Intent-based agentic orchestration in edge intelligence
Thesis event information
Date and time of the thesis defence
Place of the thesis defence
L1, Linnanmaa Campus
Topic of the dissertation
Intent-based agentic orchestration in edge intelligence
Doctoral candidate
Master of Science (Electronics Engineering) Alaa Saleh
Faculty and unit
University of Oulu Graduate School, Faculty of Information Technology and Electrical Engineering, Center for Applied Computing
Subject of study
Computer Science and Engineering
Opponent
Professor Pablo Fernandez, University of Seville
Custos
Assistant Professor Lauri Lovén, University of Oulu
Smarter, personalized, and resource-efficient artificial intelligence services that adapt to user needs and changing environments
Imagine arriving on a university campus and asking your phone to find a quiet room for a meeting. Fulfilling this simple request may require your phone to communicate with several systems across different buildings. As you move around the campus or your network connection changes, these systems may need to interact and transfer tasks to ensure that the service continues without interruption.
This example illustrates how artificial intelligence is becoming an important part of everyday environments, such as smart buildings, workplaces, and communication networks, where different devices and systems must work together to respond to users’ needs. However, these systems must respond quickly to changing user needs, manage limited computing and memory resources efficiently, and maintain continuous services.
To address these challenges, this dissertation designed ways for artificial intelligence systems to work together, manage information, and adapt to changing needs and conditions to deliver personalized and reliable services while making efficient use of available resources. The study tested these ideas in smart-building scenarios using several artificial intelligence models and data from the University of Oulu’s Smart Campus. The experiments showed that artificial intelligence systems can improve decision-making, balance response speed with decision quality, and use computing resources more efficiently by adapting to users’ goals, available resources, and changing environmental conditions.
Overall, the dissertation contributes to the development of artificial intelligence systems that are more responsive, adaptable, and efficient. The findings provide a foundation for future smart buildings, intelligent services, and communication networks that better respond to individual needs while reducing unnecessary resource use.
This example illustrates how artificial intelligence is becoming an important part of everyday environments, such as smart buildings, workplaces, and communication networks, where different devices and systems must work together to respond to users’ needs. However, these systems must respond quickly to changing user needs, manage limited computing and memory resources efficiently, and maintain continuous services.
To address these challenges, this dissertation designed ways for artificial intelligence systems to work together, manage information, and adapt to changing needs and conditions to deliver personalized and reliable services while making efficient use of available resources. The study tested these ideas in smart-building scenarios using several artificial intelligence models and data from the University of Oulu’s Smart Campus. The experiments showed that artificial intelligence systems can improve decision-making, balance response speed with decision quality, and use computing resources more efficiently by adapting to users’ goals, available resources, and changing environmental conditions.
Overall, the dissertation contributes to the development of artificial intelligence systems that are more responsive, adaptable, and efficient. The findings provide a foundation for future smart buildings, intelligent services, and communication networks that better respond to individual needs while reducing unnecessary resource use.
Created 13.11.2026 | Updated 16.11.2026