Industrial Systems Institute

Metaverse presents significant opportunities for educational advancement by facilitating immersive, personalized, and interactive learning experiences through technologies such as virtual reality (VR), augmented reality (AR), extended reality (XR), and artificial intelligence (AI). However, this potential is compromised if digital environments fail to uphold individuals’ privacy, autonomy, and equity. Despite their widespread adoption, the privacy implications of these environments remain inadequately understood, both in terms of technical vulnerabilities and legislative challenges, particularly regarding user consent management. Contemporary Metaverse systems collect highly sensitive information, including biometric signals, spatial behavior, motion patterns, and interaction data, often surpassing the granularity captured by traditional social networks. The lack of privacy-by-design solutions, coupled with the complexity of underlying technologies such as VR/AR infrastructures, 3D tracking systems, and AI-driven personalization engines, makes these platforms vulnerable to security breaches, data misuse, and opaque processing practices. This study presents a structured literature review and comparative analysis of privacy risks, consent mechanisms, and digital boundaries in metaverse platforms, with particular attention to educational contexts. We argue that privacy-aware design is essential not only for ethical compliance but also for supporting the long-term sustainability goals of digital education. Our findings aim to inform and support the development of secure, inclusive, and ethically grounded immersive learning environments by providing insights into systemic privacy and policy shortcomings.

ABSTRACT Coaxial rotor 6‐DOF UAVs can find use in several defence and civilian tasks. In this article, two different control methods are proposed for the control of this type of drones: (i) nonlinear optimal control and (ii) multi‐loop flatness‐based control. The dynamic model of the coaxial rotor drone is formulated and differential flatness properties are proven about it. To apply the nonlinear optimal control method, the dynamic model of the coaxial rotor drone undergoes approximate linearization with first‐order Taylor‐series expansion and through the computation of the associated Jacobian matrices. To apply multi‐loop flatness‐based control, the dynamic model of the UAV is decomposed in two subsystems connected in chained form. This means that the state vector of the second subsystem becomes virtual control input to the first subsystem, while the virtual control input of the first subsystem becomes a setpoints vector for the second subsystem. The two proposed control schemes ensure stabilization and precise flight‐path tracking for the coaxial rotor UAV. Both control methods avoid changes of state variables and complicated state‐space model transformations.

A new nonlinear optimal control method is proposed for solving the problem of control and stabilization of a two-link robotic manipulator where the first link is rigid and the second link is flexible. The dynamic model of this robot describes both the rotational motion of the two links and the vibration modes which are excited at the flexible link. The control and stabilization problem of this 2-DOF rigid-flexible robot is nontrivial due to complex nonlinear dynamics and underactuation. To apply the proposed nonlinear optimal control method, the dynamic model of the two-link rigid-flexible robotic manipulator undergoes first approximate linearization around a temporary operating point that is updated at each iteration of the control algorithm. The linearization takes place through first-order Taylor series expansion and through the computation of the Jacobian matrices of the flexible manipulator’s state-space description. For the approximately linearized model of the two-link rigid-flexible robot, an H-infinity feedback controller is designed. Actually, the H-infinity controller stands for the solution of the optimal control problem for the rigid-flexible robot under uncertainty and external perturbations. For the computation of the feedback gains of the H-infinity controller, an algebraic Riccati equation is solved at each time-step of the control method. The stability properties of the control algorithm are proven through Lyapunov analysis. First, it is shown that the control scheme achieves H-infinity tracking performance which signifies robustness for the control loop of the two-link rigid-flexible robot under uncertainties and external perturbations. Next, it is also shown that the control loop of the 2-DOF rigid-flexible robot is globally asymptotically stable. The proposed control method achieves fast and accurate tracking of setpoints under moderate variations of the control inputs. The concept of the proposed nonlinear optimal control method has been to approximate locally and at each sampling interval the nonlinear dynamics of the rigid-flexible robotic manipulator by an equivalent linear state-space description and to solve the H-infinity control problem for the linearized model. Finally, the control inputs are applied to the initial nonlinear dynamics and stability proof is given for the initial nonlinear state-space model.

In this paper, we study the problem of shape part retrieval in the point cloud domain. Existing shape retrieval methods typically rely on the presence of a complete query object, but what if the part of interest is missing? We present the Part Retrieval Pipeline (PReP), which combines metric learning techniques with a trained classification model to evaluate the suitability of potential replacement parts from a database, within an application scenario aimed at circular economy. Through a progressively more difficult training procedure, PReP learns to recognize suitable parts based solely on shape context. Owing to its compact parameterization and low computational requirements, it can search a repository of tens of thousands of spare parts in just a few seconds. We also establish an alternative baseline approach for comparison, document the unique challenges associated with this task, and identify key design choices to address them.

LiDAR segmentation is essential for autonomous driving, but its accuracy heavily depends on the sensor resolution. While high-resolution LiDAR offers precise environmental detail, its cost is prohibitive. Low-cost sensors, by contrast, produce sparse point clouds that degrade segmentation quality. Recent LiDAR super-resolution (SR) methods aim to address this, but face three key challenges: (1) they are optimized independently of segmentation, leading to suboptimal results; (2) they require costly post-processing to remove outliers; and (3) they introduce significant computational overhead, often surpassing the complexity of the segmentation network. To address these challenges, we propose a novel conditional optimization problem for LiDAR SR, inspired by recent advances in conditional generative models, which use external guidance signals to influence the generation process. In our case, semantic priors are extracted from the lowresolution data to guide the reconstruction toward semantically meaningful regions. This conditional formulation is efficiently solved using a model-based deep unrolling strategy, yielding a highly efficient conditional SR network. To further enhance performance, the resulting SR model is seamlessly integrated with a high-resolution segmentation network to form a unified, end-to-end trainable architecture. Experiments on public benchmarks show that our method outperforms state-of-the-art LiDAR SR approaches, particularly in preserving semantically important structures. In addition, we successfully deploy the proposed framework on our demo vehicle equipped with an NVIDIA Orin and a 16-channel LiDAR sensor, confirming its practical applicability in real-world driving environments. The code will be available at the project repository<sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">1</sup>.

The complexity of 5G and 6G networks requires advanced automation through Zero-Touch Network and Service Management (ZSM). While Intent-Based Networking (IBN) offers a promising approach, a significant bottleneck remains in translating, natural language (NL) user requirements into standardized, machine-readable intent formats. We present an adaptive intent generation and routing framework for autonomous network management, which automates the translation of NL intents into validated TM Forum (TMF) specifications. The proposed system employs a multi-stage pipeline with adaptive routing, a collection of specialized Large Language Model (LLM) workers for intent generation, and a validation subsystem(syntactic and semantic) of intents. The adaptive routing dynamically assigns user intent requests to the most appropriate LLM worker, leveraging historical performance data, task specialization, and system constraints to optimize intent generation. Experiments using lightweight, open-source LLMs confirm the system’s ability to generate TeleManagement Forum (TMF)-compliant intents reliably. The adaptive routing efficiently navigates model performance variations to optimize accuracy and resource usage, while the validation subsystem guarantees that only syntactically and semantically correct intents are produced. The proposed framework enables end-to-end automation of translating natural language requirements into validated network intents, resulting in improved reliability, accuracy, and scalability.

Therapies targeting neurodegenerative diseases via brain ventricles and spinal parenchyma face delivery challenges. Systemic administration is ineffective due to the blood-brain barrier, while direct surgical access, especially for multi-site delivery, is highly invasive. The spinal subarachnoid space offers potential for microcatheter-based delivery, but existing robotic catheter technologies are unsuitable due to spinal anatomy constraints. This paper presents a miniaturised and sensorised steerable eversion-growing robot tailored to navigation of the subarachnoid space of the spine. The property of eversion reduces interaction forces with the anatomy, rendering our approach safer than microcatheters that need to be pushed. Our system is capable of real-time tip force estimation with three degrees of freedom (DoF) using fibre Bragg gratings (FBG). Additionally, it incorporates a micro-endoscope and a steerable tip, all within a tiny 2mm outer diameter. The system’s navigation, sensing, and imaging capabilities were evaluated using a realistic up-scaled phantom of the subarachnoid space covering the cervical spine, demonstrating interaction forces within the safe range of 2-5N during phantom navigation. Comparison study of instrument-tissue interactions further approved its clinical relevance, presenting a 73.78% decrease of the mean absolute forces to traditional insertion without the sheath in global measurements.

The global agricultural sector is undergoing a transformative shift, driven by increasing food demands, climate variability and the need for sustainable practices. SUSTAINABLE is a smart farming platform designed to integrate IoT, AI, satellite imaging, and role-based task orchestration to enable efficient, traceable, and sustainable agriculture with a pilot usecase in viticulture. This paper explores current smart agriculture solutions, presents a comparative evaluation, and introduces SUSTAINABLE's key features, including satellite index integration, real-time environmental data, and role-aware task management tailored to Mediterranean vineyards.