Autonomous drone swarms offer significant potential for civil applications such as search and rescue, disaster response, and infrastructure monitoring, yet they challenge regulatory frameworks originally designed for single unmanned aircraft under direct human control. This paper analyzes key regulatory and safety issues arising from civil autonomous drone swarms, focusing on accountability, autonomy governance, airspace integration, and public safety and privacy. Using European Union law as the primary reference, complemented by a Swedish national case study and comparative insights from selected non-EU jurisdictions, the paper identifies structural gaps in aviation, AI, and data protection regulation. It argues that compliance-by-design - embedding legal and safety requirements directly into swarm architectures and operational concepts - is essential for enabling safe, lawful, and publicly acceptable deployment of autonomous drone swarms in civil airspace. © 2026 IEEE.

Passive Wi-Fi localization for critical-infrastructure security operations centers (SOCs) faces three interconnected limitations. First, many existing methods produce single-point coordinate estimates without calibrated uncertainty, making them unsuitable for automated SOC response. Second, localization pipelines often lack an evidentiary chain of custody, limiting reliable post-incident auditability. Third, SOC automation cannot safely rely on uncalibrated confidence values because erroneous high-impact actions and missed escalations carry asymmetric operational costs. This study presents a Blockchain-Enabled Uncertainty-Aware Passive Wi-Fi Localization framework for heterogeneous sensor networks composed of stationary sensors, mobile receivers, and UAV-assisted collection nodes. Instead of producing a single coordinate estimate, the method derives a posterior spatial distribution with calibrated uncertainty from monitor-mode observations, including RSSI aggregates, management/control frame features, channel occupancy indicators, and receiver logs. The framework combines three tightly coupled components: (i) Bayesian coordinate estimation with robust loss functions and range-dependent error modeling; (ii) uncertainty calibration that converts posterior confidence into operational SOC response modes (AUTO, VERIFY, and OBSERVE) via empirical coverage metrics and reliability diagrams; and (iii) a permissioned evidentiary logging layer that anchors integrity-relevant metadata and policy labels on-chain while keeping raw telemetry off-chain for tamper-evident auditability and scalability. The coupling between layers is explicit: calibrated confidence scores govern smart-contract gating conditions, and smart-contract policy thresholds feed back into the calibration stage. Field validation shows that localization performance degrades markedly beyond approximately 40 m, indicating a practical boundary for confident automated action. The proposed framework integrates passive sensing, uncertainty-aware localization, and blockchain-based evidentiary trust for secure critical-infrastructure sensor networks. Its key contributions are: (1) a posterior-distribution-based passive localization pipeline; (2) empirical coverage metrics for calibrating SOC response thresholds; (3) a hybrid on-chain/off-chain architecture linking localization outputs to a permissioned ledger; and (4) field validation establishing the 40 m operational validity boundary. © 2026 by the authors.

By 2025, distributed cloud computing services have become commonplace and widely adopted, transforming the IT landscape across industries and institutions. These services offer numerous benefits that enhance the efficiency and scalability of IT operations. With the recent surge and popularization of Artificial Intelligence image classification and image object detection workloads, cloud computing has become a popular method to train, evaluate, and serve such jobs. This article explores the AI image object detection frameworks of the three leading cloud providers-Amazon Web Services, Microsoft Azure, and Google-based on Convolutional Neural Network (CNN) architecture. It presents a practical comparative analysis of their respective solutions: Amazon Rekognition, Azure Custom Vision, and Vertex AI, alongside a locally trained object detection model using YOLO11m. The collected results, along with a discussion of the advantages, limitations, and key insights, are presented throughout the paper. © The Author(s) 2026.

This paper presents the design and development of an alternative, cost-effective automated piece positioning system, specifically tailored for Small and Medium-sized Enterprises (SMEs), which integrates computer vision with EtherCAT-controlled servo motors. The proposed method combines a robust vision system with an AI-enhanced algorithm based on edge detection to precisely identify object contours. This enables a Programmable Logic Controller (PLC) to control the servo motor, adjusting the piece’s angle with high accuracy. Experimental results demonstrate the solution’s practical viability, achieving a minimal angular oscillation of less than 0.0012° and a promising low image processing time of approximately 20ms, showcasing its potential for enhancing manufacturing efficiency and quality in industrial applications. © Augusto de Souza, Alexandre dos Santos Roque, Carlos Eduardo Pereira, and.

Agrivoltaics, the dual use of land for energy and food production, promises to improve people’s lives, but its adoption is hindered by various challenges. Drones appear well suited to supporting the enhanced need for monitoring and maintenance associated with agrivoltaic systems, due to their affordability, efficiency, and operational flexibility. A challenge is understanding what has already been done and what remains to be done in regard to the opportunities and obstacles that exist for drones in agrivoltaics, also in various local sectors, and with respect to the trend toward artificial intelligence (AI) and growing autonomy in technology. Thus, the current paper presents the results of an exploratory, big-picture review, identifying important themes and potentially interesting gaps that could be leveraged in the next years, including perspectives on research, industry, AI, and local initiatives. Finally, opportunities and roles are discussed, with the aim of stimulating interest and ideation in the area. Overall, the literature suggests that activity in the area is multinational, high, and increasing, yet still in an exploratory phase. Future work will include transferring various technical capabilities, to conduct integrated studies that leverage latest innovations in real world environments. © 2026 The Author(s). Published by Elsevier B.V. This is an open access article under the CC BY license. http://creativecommons.org/licenses/by/4.0/

The International Organization for Standardization (ISO) 7010 standard is a critical component in maintaining safety and health in various environments, including workplaces, public areas and where heavy machinery is operated. ISO 7010 prescribes safety signs across standard images for the purposes of accident prevention and hazardous protection, providing a universal language of safety, that can be understood by everyone, regardless of their nationality or native language. These symbols are unique, having only one symbol for each meaning to avoid confusion across industries. In this context, this work proposes a solution to train, evaluate and deploy an AI image detection model that is able to identify such signs in an image (frame), and paired with natural language processing (NLP) through the use of proprietary Large Language Models (LLM), be able to identify in which location of the document such signs are present. For that, the framework uses cutting-edge LLM technologies as Function Tools and RAG (Retrieval Augmented Generation) practices. Results obtained in this framework accurately retrieves isolated pictograms from semantic queries, but faces reduced classification accuracy when images contain adjacent text. This work can also bring other potential benefits for places using ISO 7010 standards, such as enhanced safety compliance, real-time monitoring, accessibility and audit. © The Author(s) 2026.

The Internet of Robotic Things (IoRT) represents the rise of a new paradigm enabling robots to serve not only as autonomous units but also as intelligent interconnected entities that can interact, collaborate, and share information through the edge, cloud and other data networks. IoRT is a technological progress and the fusion of Robotics with the Internet of Things (IoT), artificial intelligence (AI), and edge-Computing, IoRT can benefit from the next-generation spatial web, Web 4.0 (the intelligent immersive knowledge Web), by enhancing data processing, situational awareness, and integration with immersive technologies, software-defined automation (SDA), and spatial computing technologies. Semantic Web and Web 4.0 technologies are becoming common in robotics projects for exchanging data and enabling data set interoperability. The main challenge is to upgrade how robotic things interact with each other and their environment in a more situation-aware fashion, enabling IoRT situation-aware capabilities. This paper reviews the definition of IoRT considering the latest developments in sensor technology and data management systems and uses a novel survey methodology to find, classify, and reuse robotic expertise and present it to the community and engineering experts. The survey is shared through the LOV4IoT-Robotics ontology catalog, which is available online. This catalog demonstrates how best practices for data sharing and data set interoperability are also used to extract robotic knowledge semi-automatically. A set of relevant semantic-enabled projects designed by domain experts that focused on extracting robotic knowledge was included. © 2025 Wiley Periodicals LLC.

Autonomous vehicles (AVs) represent a technology with significant social and environmental benefits. By reducing dependence on the human factor, which is responsible for 94% of the 1.35 million annual traffic deaths globally, AVs have the potential to increase road safety and save lives. Complementary technologies, such as Vehicle-to-Everything (V2X) communication, further enhance traffic management, reducing congestion by up to 40% and improving energy efficiency with fuel savings of up to 15%. However, V2X systems are particularly vulnerable to cyber attacks, such as spoofing, which injects false information, disrupting the flow of traffic and compromising the safety of AVs. This paper proposes an innovative framework for detecting and mitigating spoofing attacks in V2X communications. The solution combines Direction of Arrival (DoA) estimation with advanced object detection algorithms, such as YOLOv8, to identify anomalous signals and locate malicious transmitters. By integrating Artificial Intelligence (AI) techniques, the framework makes it possible to accurately classify attackers and select customized countermeasures, ensuring greater network reliability and security. The simulation results demonstrate the framework's effectiveness in various dynamic scenarios using data from antenna arrays and camera-based object detection. In addition, they highlight the importance of sensor data fusion to improve anomaly detection accuracy, optimize decision-making processes in AVs, and enable robust cross-validation of transmitted information. © 2026 The Authors.

This paper presents an autonomous Unmanned Aerial Vehicle (UAV)-based system for cleaning glass building facades. The proposed solution combines a real-time computer vision subsystem, built upon the You Only Look Once (YOLO) object detection architecture for accurate window localization, with an embedded control module that actuates high-pressure water nozzles to perform the cleaning task. A custom dataset composed of both real-world and synthetic images of glass facades was built and used to train and evaluate several YOLO models. Among them, YOLOv12 achieved the highest performance, attaining a mean Average Precision (mAP50) of 80% for the glass window class. The integrated system was thoroughly validated through simulations in Gazebo/ROS2 and real-world field experiments using a dedicated UAV platform (Skyclean drone), demonstrating reliable detection performance and safe operation. Results confirm that the approach effectively automates facade maintenance, significantly improving cleaning operations, reducing costs and human risk exposure, and providing a robust, extensible framework for future fully autonomous facade cleaning systems. © 2028 World Scientific Publishing Company.

Unmanned aerial platforms (UAPs) have emerged as a flexible alternative to fixed infrastructure for supporting wireless sensing and vehicular applications, offering improved line-of-sight conditions and dynamic deployment capabilities. This work investigates an aerial platform-enabled bistatic multiple-in multiple-out (MIMO) radar architecture applied to Vehicular Ad Hoc Networks (VANETs), where an airborne node operates as an aerial Roadside Unit (RSU) to support sensing and localization of ground vehicles. By exploiting the inherent spatial separation between transmitter and receiver in bistatic configurations, the proposed framework enhances geometric diversity compared to conventional monostatic and terrestrial RSU-based sensing approaches. A signal model for the aerial-assisted bistatic MIMO radar scenario is formulated, and tensor-based signal processing techniques are employed to jointly estimate the direction of departure (DoD) and direction of arrival (DoA) of vehicle targets. The multidimensional structure of the received signals is naturally captured using tensor representations, enabling efficient parameter estimation with automatic angle pairing. Numerical simulations are conducted to assess the impact of the aerial RSU on angular estimation performance under different signal-to-noise ratio conditions. The results indicate that the proposed aerial platform-enabled bistatic MIMO radar framework achieves angular estimation performance comparable to that of fixed infrastructure while providing increased flexibility and adaptability, highlighting its potential as a complementary sensing solution for future intelligent transportation systems. © 2026 Warsaw University of Technology.