Paper title: EVLM: Intent-Driven Edge Vision Language Model for UAV-Based Power Line Inspection

Authors: Reza Farahani (DSG, TU Wien, Austria), Zoha Azimi (Christian Doppler Laboratory ATHENA, ITEC, University of Klagenfurt, Austria), Ilir Murturi (Department of Mechatronics, University of Prishtina, Kosova), Arda Goknil (SINTEF, Oslo, Norway), Sagar Sen (SINTEF, Oslo, Norway), Christian Timmerer (Christian Doppler Laboratory ATHENA, ITEC, University of Klagenfurt, Austria), Schahram Dustdar (DSG, TU Wien, Austria)

Conference: 2026 IEEE International Conference on Edge Computing and Communications (IEEE EDGE 2026)

Abstract: 

Inspection of critical infrastructure, such as power lines, is increasingly conducted using unmanned aerial vehicles (UAVs) that capture aerial video for subsequent human review. Although recent edge-based approaches deploy onboard object
detectors to identify predefined defect classes, these pipelines remain closed-set, task-specific, and largely decoupled from operator intent and edge resource constraints. This paper introduces EVLM, an intent-driven vision-language framework for onboard UAV-based power line inspection. Given a high-level operator intent, EVLM (i) leverages lightweight histogram-based frame filtering to extract salient key frames under bounded compute budgets, (ii) executes a domain-adapted vision language model (VLM) directly on the UAV for intent-conditioned multimodal reasoning, and (iii) synthesizes structured inspection reports together with a minimal set of evidence frames, replacing continuous raw video transmission with compact semantic outputs. To align the VLM with infrastructure inspection semantics while preserving edge efficiency, we perform parameter-efficient fine-tuning using Low-Rank Adaptation (LoRA), enabling domain specialization without updating the full model parameters. We implement and fully deploy EVLM on an NVIDIA Jetson device representative of UAV-class onboard hardware and evaluate it using 20 publicly released power line inspection video sequences spanning 8 heterogeneous environments and 5 operational intent categories. Experimental results show a data reduction of 94.8 %, with transmitted data decreasing from 485 kB to 25 kB per 4 s segment, corresponding to 72.75 MB versus 3.75 MB over a 10 min inspection mission. EVLM operates feasibly on embedded hardware, maintaining moderate CPU/GPU utilization and bounded power consumption (5.6 W), while producing interpretable, intent-aligned inspection outputs with richer semantic insights than detection-centric baselines.

Quantifying Inter-City Network Latency in Europe: A Measurement based Study for Time-Critical Cloud Services

3rd Workshop on Engineering Techniques for Distributed Computing Continuum Systems (EDCCS), 22-25 June 2026, Seoul, South Korea

Authors: Thomas Schleicher, Kurt Horvath, Dragi Kimovski, Bernd Spiess, Oliver Hohlfeld

Abstract: Time-critical cloud and edge services depend on predictable and low-latency wide-area connectivity, yet inter-city network behavior often deviates from expectations based on geographic distance alone. This paper presents an evaluation framework and results on inter-city network latency across major European metropolitan areas, treating latency as a non-functional property relevant to benchmarking and service placement in cloud computing. We develop a scalable measurement framework based on a distributed probing infrastructure, analyze round-trip latency, and assess spatial efficiency and temporal stability. Initial results reveal unexpectedly high latency on long-distance paths from the Iberian Peninsula toward Turkey. Distance-normalized analysis further exposes pronounced inefficiencies on short-distance paths between Greece and Turkey, suggesting non-distance-related network effects beyond geographic proximity. Temporal analysis shows elevated latency variance and instability on paths involving Turkey, while most other inter-city connections closely follow distance-based expectations and remain stable over time. These findings highlight the importance of distance-normalized and stability-aware metrics for evaluating wide-area cloud connectivity. The presented methodology and results provide practical insight for benchmarking, placement, and operation of latency-sensitive cloud services across geographically distributed infrastructures.