The selected paper, titled “Graph-Based Decentralized Exploration and Semantic Inspection for Multi-Robot Systems,” is the result of the Master’s thesis by Nada Abbas, carried out at the University of Zagreb, Faculty of Electrical Engineering and Computing (FER).
The research, developed within the framework of the Erasmus Mundus Master in Intelligent Field Robotic Systems (IFRoS), has been accepted for publication at the 2025 International Conference on Unmanned Aircraft Systems (ICUAS), a leading international venue for cutting-edge advancements in autonomous aerial systems.
The paper introduces a unified framework that enables a team of unmanned aerial vehicles (UAVs) to explore and inspect complex environments—such as tunnels and underground infrastructures—without centralized coordination. At its core lies a decentralized, graph-based planning strategy that brings together real-time task allocation, volumetric mapping, and semantic object inspection.
To enhance efficiency and reduce communication load, the UAVs exchange only essential data such as high-gain frontiers and designated “No-Go Zones.” A key innovation of the system is its ability to autonomously switch between exploration and inspection modes, with dynamic task reassignment ensuring mission goals are met within time and energy constraints. The result is a scalable, robust approach to multi-robot collaboration in challenging environments.
Nada Abbas completed her studies in the IFRoS program and now serves as AI Team Lead at Avemoy FlexCo in Austria. In her current role, she leads the development of AI-powered systems for autonomous agricultural monitoring, leveraging UAV platforms and edge computing technologies.
Her research exemplifies the type of high-impact, application-driven work fostered by the IFRoS program—an international initiative that promotes academic excellence and innovation through collaboration among Europe’s top institutions in robotics and artificial intelligence.