Papers

Integrated sensing and communication (ISAC) has emerged as a key enabler for sixth-generation (6G) wireless networks, supporting spectrum sharing and hardware integration. Beyond communication enhancement, ISAC also enables high-accuracy environment reconstruction and imaging, which are crucial for applications such as autonomous driving and digital twins. This paper proposes a 4D imaging framework fully compliant with the 5G New Radio (NR) protocol, ensuring compatibility with cellular systems. Specifically, we develop an end-to-end processing chain that covers waveform generation, echo processing, and multi-BS point cloud fusion. Furthermore, we introduce Zoom-OMP, a coarse-to-fine sparse recovery algorithm for high-resolution angle estimation that achieves high accuracy with reduced computational cost. The simulation results demonstrate that the proposed framework achieves robust 4D imaging performance with superior spatial accuracy and reconstruction quality compared to conventional benchmarks, paving the way for practical ISAC-enabled environment reconstruction in 6G networks.

Benefiting from the rapid development of metamaterials and metasurfaces, the holographic multiple-input and multiple-output surface (HMIMOS) has been regarded as a promising solution for future wireless networks recently. By densely packing numerous radiation elements together, HMIMOS is capable of realizing accurate beamforming with low hardware complexity. However, enormous radiation elements on the HMIMOS lead to high computational complexity and signaling overhead when applying traditional beamforming schemes relying on instantaneous channel state information (CSI). To tackle this problem, we propose a two-timescale optimization scheme to minimize the required transmission power under the constraint of all users' quality-of-service (QoS). Specifically, the beampatterns at the base station (BS) and the user equippment (UE) are optimized over the slowly changing statistical CSI based on the constrained stochastic successive convex approximation (CSSCA) algorithm. Then, the instantaneous CSI is utilized to design the precoding matrix in order to ensure the system performance without significant increase in computational cost, due to the small number of feeds on the HMIMOS. Simulation results demonstrate the effectiveness of our proposed method compared to other baselines.

Nuclei segmentation is the cornerstone task in histology image reading, shedding light on the underlying molecular patterns and leading to disease or cancer diagnosis. Yet, it is a laborious task that requires expertise from trained physicians. The large nuclei variability across different organ tissues and acquisition processes challenges the automation of this task. On the other hand, data annotations are expensive to obtain, and thus, Deep Learning (DL) models are challenged to generalize to unseen organs or different domains. This work proposes Local-to-Global NuSegHop (LG-NuSegHop), a self-supervised pipeline developed on prior knowledge of the problem and molecular biology. There are three distinct modules: (1) a set of local processing operations to generate a pseudolabel, (2) NuSegHop a novel data-driven feature extraction model and (3) a set of global operations to post-process the predictions of NuSegHop. Notably, even though the proposed pipeline uses { no manually annotated training data} or domain adaptation, it maintains a good generalization performance on other datasets. Experiments in three publicly available datasets show that our method outperforms other self-supervised and weakly supervised methods while having a competitive standing among fully supervised methods. Remarkably, every module within LG-NuSegHop is transparent and explainable to physicians.

Multi-drone surveillance systems offer enhanced coverage and robustness for pedestrian tracking, yet existing approaches struggle with dynamic camera positions and complex occlusions. This paper introduces MATRIX (Multi-Aerial TRacking In compleX environments), a comprehensive dataset featuring synchronized footage from eight drones with continuously changing positions, and a novel deep learning framework for multi-view detection and tracking. Unlike existing datasets that rely on static cameras or limited drone coverage, MATRIX provides a challenging scenario with 40 pedestrians and a significant architectural obstruction in an urban environment. Our framework addresses the unique challenges of dynamic drone-based surveillance through real-time camera calibration, feature-based image registration, and multi-view feature fusion in bird's-eye-view (BEV) representation. Experimental results demonstrate that while static camera methods maintain over 90% detection and tracking precision and accuracy metrics in a simplified MATRIX environment without an obstruction, 10 pedestrians and a much smaller observational area, their performance significantly degrades in the complex environment. Our proposed approach maintains robust performance with $\sim$90% detection and tracking accuracy, as well as successfully tracks $\sim$80% of trajectories under challenging conditions. Transfer learning experiments reveal strong generalization capabilities, with the pretrained model achieving much higher detection and tracking accuracy performance compared to training the model from scratch. Additionally, systematic camera dropout experiments reveal graceful performance degradation, demonstrating practical robustness for real-world deployments where camera failures may occur. The MATRIX dataset and framework provide essential benchmarks for advancing dynamic multi-view surveillance systems.

This paper explores a joint optimization of transmit power allocation and radiation coefficients in a downlink Pinching Antenna SyStem (PASS) employing Non-Orthogonal Multiple Access (NOMA). By leveraging the PASS-enabled flexible channel adjustment and NOMA's power allocation adaptability, a sum rate maximization problem is formulated with the objective of simultaneously optimizing base station (BS)'s transmit power coefficients and pinching antenna (PA)'s radiation powers. Due to its non-convexity and complexity, the formulated optimization problem is challenging to solve directly. Hence, we decompose the main problem into two sub-problems, namely transmit power allocation sub-problem and PA radiation power allocation sub-problem. In the first sub-problem, closed-form solutions are derived for the BS's power allocation among NOMA users. Meanwhile, in the second sub-problem, we optimize the PA's radiation power utilizing successive convex approximation (SCA). These two sub-problems are solved alternatively using Alternating Optimization (AO) until convergence. It should be noted that decoding order plays a significant role in NOMA-assisted PASS. Hence, two variations of decoding order are considered, namely: i) a high-complexity exhaustive search approach, and, ii) a low-complexity alternative that utilizes pre-determined channel information. Numerical results show that our proposed approach substantially improves the system's sum-rate compared to widely adopted equal power allocation PASS schemes.

Due to recent blackout and system split incidents in power grids worldwide, as well as increased system complexity in view of the energy transition, there has been increasing interest in re-evaluating existing Power System Restoration (PSR) plans. In restoration scenarios, due to low island inertia, it is necessary to ensure not only the static, but also the dynamic stability of the system. In this paper, we pose and solve a formulation of the optimal PSR problem including frequency dynamics. We validate the switching constraints for global optimality within a static version of the formulation using a brute-force tree search method. We apply the dynamic problem formulation to the IEEE 9-Bus model, and show that the optimal switching sequence using the static formulation would violate dynamic constraints, illustrating the importance of dynamic considerations in PSR planning.

Hybrid power plants (HPPs) combine multiple power generators (conventional/variable) and energy storage capabilities to support generation inadequacy and grid demands. This paper introduces a modeling and control design framework for hybrid power plants (HPPs) consisting of a wind farm, solar plant, and battery storage. Specifically, this work adapts established modeling paradigms for wind farms, solar plants and battery models into a control affine form suitable for control design at the supervisory level. In the case of wind and battery models, generator torque and cell current control laws are developed using nonlinear control and control barrier function techniques to track a command from a supervisory control law while maintaining safe and stable operation. The utility of this modeling and control framework is illustrated through a test case using a utility demand signal for tracking, time varying wind and irradiance data, and a rule-based supervisory control law.

This paper presents a switch-type attenuator working from 20 to 100 GHz. The attenuator adopts a capacitive compensation technique to reduce phase error. The small resistors in this work are implemented with metal lines to reduce the intrinsic parasitic capacitance, which helps minimize the amplitude and phase errors over a wide frequency range. Moreover, the utilization of metal lines also reduces the chip area. In addition, a continuous tuning attenuation unit is employed to improve the overall attenuation accuracy of the attenuator. The passive attenuator is designed and fabricated in a standard 65nm CMOS. The measurement results reveal a relative attenuation range of 7.5 dB with a continuous tuning step within 20-100 GHz. The insertion loss is 1.6-3.8 dB within the operation band, while the return losses of all states are better than 11.5 dB. The RMS amplitude and phase errors are below 0.15 dB and 1.6{\deg}, respectively.

We consider a time-slotted job-assignment system with a central server, N users and a machine which changes its state according to a Markov chain (hence called a Markov machine). The users submit their jobs to the central server according to a stochastic job arrival process. For each user, the server has a dedicated job queue. Upon receiving a job from a user, the server stores that job in the corresponding queue. When the machine is not working on a job assigned by the server, the machine can be either in internally busy or in free state, and the dynamics of these states follow a binary symmetric Markov chain. Upon sampling the state information of the machine, if the server identifies that the machine is in the free state, it schedules a user and submits a job to the machine from the job queue of the scheduled user. To maximize the number of jobs completed per unit time, we introduce a new metric, referred to as the age of job completion. To minimize the age of job completion and the sampling cost, we propose two policies and numerically evaluate their performance. For both of these policies, we find sufficient conditions under which the job queues will remain stable.

In this paper, we focus on recovery control of nonlinear systems from attacks or failures. The main challenges of this problem lie in (1) learning the unknown dynamics caused by attacks or failures with formal guarantees, and (2) finding the invariant set of states to formally ensure the state deviations allowed from the nominal trajectory. To solve this problem, we propose to apply the Recurrent Equilibrium Networks (RENs) to learn the unknown dynamics using the data from the real-time system states. The input-output property of this REN model is guaranteed by incremental integral quadratic constraints (IQCs). Then, we propose a funnel-based control method to achieve system recovery from the deviated states. In particular, a sufficient condition for nominal trajectory stabilization is derived together with the invariant funnels along the nominal trajectory. Eventually, the effectiveness of our proposed control method is illustrated by a simulation example of a DC microgrid control application.

Recent breakthroughs in language-queried audio source separation (LASS) have shown that generative models can achieve higher separation audio quality than traditional masking-based approaches. However, two key limitations restrict their practical use: (1) users often require operations beyond separation, such as sound removal; and (2) relying solely on text prompts can be unintuitive for specifying sound sources. In this paper, we propose PromptSep to extend LASS into a broader framework for general-purpose sound separation. PromptSep leverages a conditional diffusion model enhanced with elaborated data simulation to enable both audio extraction and sound removal. To move beyond text-only queries, we incorporate vocal imitation as an additional and more intuitive conditioning modality for our model, by incorporating Sketch2Sound as a data augmentation strategy. Both objective and subjective evaluations on multiple benchmarks demonstrate that PromptSep achieves state-of-the-art performance in sound removal and vocal-imitation-guided source separation, while maintaining competitive results on language-queried source separation.

Remote monitoring of cardiovascular diseases plays an essential role in early detection of abnormal cardiac function, enabling timely intervention, improved preventive care, and personalized patient treatment. Abnormalities in the heart sounds can be detected automatically via computer-assisted decision support systems, and used as the first-line screening tool for detection of cardiovascular problems, or for monitoring the effects of treatments and interventions. We propose in this paper CardioPHON, an integrated heart sound quality assessment and classification tool that can be used for screening of abnormal cardiac function from phonocardiogram recordings. The model is pretrained in a self-supervised fashion on a collection of six small- and mid-sized heart sound datasets, enables automatic removal of low quality recordings to ensure that subtle sounds of heart abnormalities are not misdiagnosed, and provides a state-of-the-art performance for the heart sound classification task. The multimodal model that combines audio and socio-demographic features demonstrated superior performance, achieving the best ranking on the official leaderboard of the 2022 George B. Moody PhysioNet heart sound challenge, whereas the unimodal model, that is based only on phonocardiogram recordings, holds the first position among the unimodal approaches (a total rank 4), surpassing the models utilizing multiple modalities. CardioPHON is the first publicly released pretrained model in the domain of heart sound recordings, facilitating the development of data-efficient artificial intelligence models that can generalize to various downstream tasks in cardiovascular diagnostics.

Landmark Inertial Simultaneous Localisation and Mapping (LI-SLAM) is the problem of estimating the locations of landmarks in the environment and the robot's pose relative to those landmarks using landmark position measurements and measurements from Inertial Measurement Unit (IMU). This paper proposes a nonlinear observer for LI-SLAM posed in continuous time and analyses the observer in a base space that encodes all the observable states of LI-SLAM. The local exponential stability and almost-global asymptotic stability of the error dynamics in base space is established in the proof section and validated using simulations.

Fluorescence Molecular Tomography (FMT) is a promising technique for non-invasive 3D visualization of fluorescent probes, but its reconstruction remains challenging due to the inherent ill-posedness and reliance on inaccurate or often-unknown tissue optical properties. While deep learning methods have shown promise, their supervised nature limits generalization beyond training data. To address these problems, we propose $\mu$NeuFMT, a self-supervised FMT reconstruction framework that integrates implicit neural-based scene representation with explicit physical modeling of photon propagation. Its key innovation lies in jointly optimize both the fluorescence distribution and the optical properties ($\mu$) during reconstruction, eliminating the need for precise prior knowledge of tissue optics or pre-conditioned training data. We demonstrate that $\mu$NeuFMT robustly recovers accurate fluorophore distributions and optical coefficients even with severely erroneous initial values (0.5$\times$ to 2$\times$ of ground truth). Extensive numerical, phantom, and in vivo validations show that $\mu$NeuFMT outperforms conventional and supervised deep learning approaches across diverse heterogeneous scenarios. Our work establishes a new paradigm for robust and accurate FMT reconstruction, paving the way for more reliable molecular imaging in complex clinically related scenarios, such as fluorescence guided surgery.

Affine Frequency Division Multiplexing (AFDM) has been proposed as an effective waveform for achieving the full diversity of doubly-dispersive (delay-Doppler) channels. While this property is closely related to range and velocity estimation in sensing, this article focuses on other AFDM features that are particularly relevant for addressing two challenges in integrated sensing and communication (ISAC) systems: (1) maintaining receiver complexity and energy consumption at acceptable levels while supporting the large bandwidths required for high delay/range resolution, and (2) mitigating interference in multiradar environments. In monostatic sensing, where direct transmitter-receiver leakage is a major impairment, we show that AFDM-based ISAC receivers can address the first challenge through their compatibility with low-complexity self-interference cancellation (SIC) schemes and reduced sampling rates via analog dechirping. In bistatic sensing, where such analog solutions may not be feasible, we demonstrate that AFDM supports sub-Nyquist sampling without requiring hardware modifications while preserving delay resolution. Finally, we show that the second challenge can be addressed by leveraging the resource-assignment flexibility of the discrete affine Fourier transform (DAFT) underlying the AFDM waveform.

Phase-shifted carrier pulse-width modulation (PSC-PWM) is a widely adopted scheduling algorithm in cascaded bridge converters, modular multilevel converters, and reconfigurable batteries. However, non-uniformed pulse widths for the modules with fixed phase shift angles lead to significant ripple current and output-voltage distortion. Voltage uniformity instead would require optimization of the phase shifts of the individual carriers. However, the computational burden for such optimization is beyond the capabilities of any simple embedded controller. This paper proposes a neural network that emulates the behavior of an instantaneous optimizer with significantly reduced computational burden. The proposed method has the advantages of stable performance in predicting the optimum phase-shift angles under balanced battery modules with non-identical modulation indices without requiring extensive lookup tables, slow numerical optimization, or complex controller tuning. With only one (re)training session for any specified number of modules, the proposed method is readily adaptable to different system sizes. Furthermore, the proposed framework also includes a simple scaling strategy that allows a neural network trained for fewer modules to be reused for larger systems by grouping modules and adjusting their phase shifts. The scaling strategy eliminates the need for retraining. Large-scale assessment, simulations, and experiments demonstrate that, on average, the proposed approach can reduce the current ripple and the weighted total harmonic distortion by up to 50 % in real time and is 100 to 500 thousand times faster than a conventional optimizer (e.g., genetic algorithms), making it the only solution for an online application.

In this study, we present and validate an ensemble-based Hankel Dynamic Mode Decomposition with control (HDMDc) for uncertainty-aware seakeeping predictions of a high-speed catamaran, namely the Delft 372 model. Experimental measurements (time histories) of wave elevation at the longitudinal center of gravity, heave, pitch, notional flight-deck velocity, notional bridge acceleration, and total resistance were collected from irregular wave basin tests on a 1:33.3 scale replica of the Delft 372 model under sea state 5 conditions at Fr = 0.425, and organized into training, validation, and test sets. The HDMDc algorithm constructs an equation-free linear reduced-order model of the seakeeping vessel by augmenting states and inputs with their time-lagged copies to capture nonlinear and memory effects. Two ensembling strategies, namely Bayesian HDMDc (BHDMDc), which samples hyperparameters considered stochastic variables with prior distribution to produce posterior mean forecasts with confidence intervals, and Frequentist HDMDc (FHDMDc), which aggregates multiple model obtained over data subsets, are compared in providing seakeeping prediction and uncertainty quantification. The FHDMDc approach is found to improve the accuracy of the predictions compared to the deterministic counterpart, also providing robust uncertainty estimation; whereas the application of BHDMDc to the present test case is not found beneficial in comparison to the deterministic model. FHDMDc-derived probability density functions for the motions closely match both experimental data and URANS results, demonstrating reliable and computationally efficient seakeeping prediction for design and operational support.

Nonlinear dynamical systems with input delays pose significant challenges for prediction, estimation, and control due to their inherent complexity and the impact of delays on system behavior. Traditional linear control techniques often fail in these contexts, necessitating innovative approaches. This paper introduces a novel approach to approximate the Koopman operator using an LSTM-enhanced Deep Koopman model, enabling linear representations of nonlinear systems with time delays. By incorporating Long Short-Term Memory (LSTM) layers, the proposed framework captures historical dependencies and efficiently encodes time-delayed system dynamics into a latent space. Unlike traditional extended Dynamic Mode Decomposition (eDMD) approaches that rely on predefined dictionaries, the LSTM-enhanced Deep Koopman model is dictionary-free, which mitigates the problems with the underlying dynamics being known and incorporated into the dictionary. Quantitative comparisons with extended eDMD on a simulated system demonstrate highly significant performance gains in prediction accuracy in cases where the true nonlinear dynamics are unknown and achieve comparable results to eDMD with known dynamics of a system.

Reconfigurable Intelligent Surfaces (RIS) have been recognized as a promising technology to enhance both communication and sensing performance in integrated sensing and communication (ISAC) systems for future 6G networks. However, existing RIS optimization methods for improving ISAC performance are mainly based on semidefinite relaxation (SDR) or iterative algorithms. The former suffers from high computational complexity and limited scalability, especially when the number of RIS elements becomes large, while the latter yields suboptimal solutions whose performance depends on initialization. In this work, we introduce a lightweight RIS phase design framework that provides a closed-form solution and explicitly accounts for the trade-off between communication and sensing, as well as proportional beam gain distribution toward multiple sensing targets. The key idea is to partition the RIS configuration into two parts: the first part is designed to maximize the communication performance, while the second introduces small perturbations to generate multiple beams for multi-target sensing. Simulation results validate the effectiveness of the proposed approach and demonstrate that it achieves performance comparable to SDR but with significantly lower computational complexity.

This paper presents a deep Koopman-based Economic Model Predictive Control (EMPC) for efficient operation of a laboratory-scale pasteurization unit (PU). The method uses Koopman operator theory to transform the complex, nonlinear system dynamics into a linear representation, enabling the application of convex optimization while representing the complex PU accurately. The deep Koopman model utilizes neural networks to learn the linear dynamics from experimental data, achieving a 45% improvement in open-loop prediction accuracy over conventional N4SID subspace identification. Both analyzed models were employed in the EMPC formulation that includes interpretable economic costs, such as energy consumption, material losses due to inadequate pasteurization, and actuator wear. The feasibility of EMPC is ensured using slack variables. The deep Koopman EMPC and N4SID EMPC are numerically validated on a nonlinear model of multivariable PU under external disturbance. The disturbances include feed pump fail-to-close scenario and the introduction of a cold batch to be pastuerized. These results demonstrate that the deep Koopmand EMPC achieves a 32% reduction in total economic cost compared to the N4SID baseline. This improvement is mainly due to the reductions in material losses and energy consumption. Furthermore, the steady-state operation via Koopman-based EMPC requires 10.2% less electrical energy. The results highlight the practical advantages of integrating deep Koopman representations with economic optimization to achieve resource-efficient control of thermal-intensive plants.

Music editing has emerged as an important and practical area of artificial intelligence, with applications ranging from video game and film music production to personalizing existing tracks according to user preferences. However, existing models face significant limitations, such as being restricted to editing synthesized music generated by their own models, requiring highly precise prompts, or necessitating task-specific retraining, thus lacking true zero-shot capability. leveraging recent advances in rectified flow and diffusion transformers, we introduce MusRec, a zero-shot text-to-music editing model capable of performing diverse editing tasks on real-world music efficiently and effectively. Experimental results demonstrate that our approach outperforms existing methods in preserving musical content, structural consistency, and editing fidelity, establishing a strong foundation for controllable music editing in real-world scenarios.

Supervisory controllers control cyber-physical systems to ensure their correct and safe operation. Synthesis-based engineering (SBE) is an approach to largely automate their design and implementation. SBE combines model-based engineering with computer-aided design, allowing engineers to focus on 'what' the system should do (the requirements) rather than 'how' it should do it (design and implementation). In the Eclipse Supervisory Control Engineering Toolkit (ESCET) open-source project, a community of users, researchers, and tool vendors jointly develop a toolkit to support the entire SBE process, particularly through the CIF modeling language and tools. In this paper, we first provide a description of CIF's symbolic supervisory controller synthesis algorithm, and thereby include aspects that are often omitted in the literature, but are of great practical relevance, such as the prevention of runtime errors, handling different types of requirements, and supporting input variables (to connect to external inputs). Secondly, we introduce and describe CIF's benchmark models, a collection of 23 freely available industrial and academic models of various sizes and complexities. Thirdly, we describe recent improvements between ESCET versions v0.8 (December 2022) and v4.0 (June 2024) that affect synthesis performance, evaluate them on our benchmark models, and show the current practical synthesis performance of CIF. Fourthly, we briefly look at multi-level synthesis, a non-monolithic synthesis approach, evaluate its gains, and show that while it can help to further improve synthesis performance, further performance improvements are still needed to synthesize complex models.

In this study, we examine the potential of high-resolution forest mapping using L-band interferometric time series datasets and deep learning modeling. Our SAR data are represented by a time series of nine ALOS-2 PALSAR-2 dual-pol SAR images acquired at near-zero spatial baseline over a study site in Asturias, Northern Spain. Reference data are collected using airborne laser scanning. We examine the performance of several candidate deep learning models from UNet-family with various combinations of input polarimetric and interferometric features. In addition to basic Vanilla UNet, attention reinforced UNet model with squeeze-excitation blocks (SeU-Net) and advanced UNet model with nested structure and skip pathways are used. Studied features include dual pol interferometric observables additionally incorporating model-based derived measures. Results show that adding model-based inverted InSAR features or InSAR coherence layers improves retrieval accuracy compared to using backscatter intensity only. Use of attention mechanisms and nested connection fusion provides better predictions than using Vanilla UNet or traditional machine learning methods. Forest height retrieval accuracies range between 3.1-3.8 m (R2 = 0.45–0.55) at 20 m resolution when only intensity data are used, and improve to less than 2.8 m when both intensity and interferometric coherence features are included. At 40 m and 60 m resolution, retrieval performance further improves, primarily due to higher SNR in both the intensity and interferometric layers. When using intensity at 60 m resolution, best achieved RMSE is 2.2 m, while when using all suitable input features the achieved error is 1.95 m. We recommend this hybrid approach for L-band SAR retrievals also suitable for NISAR and future ROSE-L missions.

Human action recognition (HAR) with multi-modal inputs (RGB-D, skeleton, point cloud) can achieve high accuracy but typically relies on large labeled datasets and degrades sharply when sensors fail or are noisy. We present Robust Cross-Modal Contrastive Learning (RCMCL), a self-supervised framework that learns modality-invariant representations and remains reliable under modality dropout and corruption. RCMCL jointly optimizes (i) a cross-modal contrastive objective that aligns heterogeneous streams, (ii) an intra-modal self-distillation objective that improves view-invariance and reduces redundancy, and (iii) a degradation simulation objective that explicitly trains models to recover from masked or corrupted inputs. At inference, an Adaptive Modality Gating (AMG) network assigns data-driven reliability weights to each modality for robust fusion. On NTU RGB+D 120 (CS/CV) and UWA3D-II, RCMCL attains state-of-the-art accuracy in standard settings and exhibits markedly better robustness: under severe dual-modality dropout it shows only an 11.5% degradation, significantly outperforming strong supervised fusion baselines. These results indicate that self-supervised cross-modal alignment, coupled with explicit degradation modeling and adaptive fusion, is key to deployable multi-modal HAR.

This paper explores the application of data-driven system identification techniques in the frequency domain to obtain simplified, control-oriented models of photosynthesis regulation under oscillating light conditions. In-silico datasets are generated using simulations of the physics-based Basic DREAM Model (BDM) Funete et al.[2024], with light intensity signals – comprising DC (static) and AC (modulated) components as input and chlorophyll fluorescence (ChlF) as output. Using these data, the Best Linear Approximation (BLA) method is employed to estimate second-order linear time-invariant (LTI) transfer function models across different operating conditions defined by DC levels and modulation frequencies of light intensity. Building on these local models, a Linear Parameter-Varying (LPV) representation is constructed, in which the scheduling parameter is defined by the DC values of the light intensity, providing a compact state-space representation of the system dynamics.

The recent and ongoing expansion of marine infrastructure, including offshore wind farms, oil and gas platforms, artificial islands, and aquaculture facilities, highlights the need for effective monitoring systems. The development of robust models for offshore infrastructure detection relies on comprehensive, balanced datasets, but falls short when samples are scarce, particularly for underrepresented object classes, shapes, and sizes. By training deep learning-based YOLOv10 object detection models with a combination of synthetic and real Sentinel-1 satellite imagery acquired in the fourth quarter of 2023 from four regions (Caspian Sea, South China Sea, Gulf of Guinea, and Coast of Brazil), this study investigates the use of synthetic training data to enhance model performance. We evaluated this approach by applying the model to detect offshore platforms in three unseen regions (Gulf of Mexico, North Sea, Persian Gulf) and thereby assess geographic transferability. This region-holdout evaluation demonstrated that the model generalises beyond the training areas. In total, 3,529 offshore platforms were detected, including 411 in the North Sea, 1,519 in the Gulf of Mexico, and 1,593 in the Persian Gulf. The model achieved an F1 score of 0.85, which improved to 0.90 upon incorporating synthetic data. We analysed how synthetic data enhances the representation of unbalanced classes and overall model performance, taking a first step toward globally transferable detection of offshore infrastructure. This study underscores the importance of balanced datasets and highlights synthetic data generation as an effective strategy to address common challenges in remote sensing, demonstrating the potential of deep learning for scalable, global offshore infrastructure monitoring.

Integrating flexible loads and storage systems into the residential sector contributes to the alignment of volatile renewable generation with demand. Besides batteries serving as a short-term storage solution, residential buildings can benefit from a Hydrogen (H2) storage system, allowing seasonal shifting of renewable energy. However, as the initial costs of H2 systems are high, coupling a Fuel Cell (FC) with a Heat Pump (HP) can contribute to the size reduction of the H2 system. The present study develops a Comfort-Oriented Energy Management System for Residential Buildings (ComEMS4Build) comprising Photovoltaics (PV), Battery Energy Storage System (BESS), and H2 storage, where FC and HP are envisioned as complementary technologies. The fuzzy-logic-based ComEMS4Build is designed and evaluated over a period of 12 weeks in winter for a family household building in Germany using a semi-synthetic modeling approach. The Rule-Based Control (RBC), which serves as a lower benchmark, is a scheduler designed to require minimal inputs for operation. The Model Predictive Control (MPC) is intended as a cost-optimal benchmark with an ideal forecast. The results show that ComEMS4Build, similar to MPC, does not violate the thermal comfort of occupants in 10 out of 12 weeks, while RBC has a slightly higher median discomfort of 0.68 Kh. The ComEMS4Build increases the weekly electricity costs by 12.06 EUR compared to MPC, while RBC increases the weekly costs by 30.14 EUR. The ComEMS4Build improves the Hybrid Energy Storage System (HESS) utilization and energy exchange with the main grid compared to the RBC. However, when it comes to the FC operation, the RBC has an advantage, as it reduces the toggling counts by 3.48% and working hours by 7.59% compared to MPC...

Brain-computer interfaces (BCIs) allow direct communication between the brain and external devices, frequently using electroencephalography (EEG) to record neural activity. Dimensionality reduction and structured regularization are essential for effectively classifying task-related brain signals, including event-related potentials (ERPs) and motor imagery (MI) rhythms. Current tensor-based approaches, such as Tucker and PARAFAC decompositions, often lack the flexibility needed to fully capture the complexity of EEG data. This study introduces Block-Term Tensor Discriminant Analysis (BTTDA): a novel tensor-based and supervised feature extraction method designed to enhance classification accuracy by providing flexible multilinear dimensionality reduction. Extending Higher Order Discriminant Analysis (HODA), BTTDA uses a novel and interpretable forward model for HODA combined with a deflation scheme to iteratively extract discriminant block terms, improving feature representation for classification. BTTDA and a sum-of-rank-1-terms variant PARAFACDA were evaluated on publicly available ERP (second-order tensors) and MI (third-order tensors) EEG datasets from the MOABB benchmarking framework. Benchmarking revealed that BTTDA and PARAFACDA significantly outperform the traditional HODA method in ERP decoding, resulting in state-of-the art performance (ROC-AUC = 91.25%). For MI, decoding results of HODA, BTTDA and PARAFACDA were subpar, but BTTDA still significantly outperformed HODA (64.52% > 61.00%). The block-term structure of BTTDA enables interpretable and more efficient dimensionality reduction without compromising discriminative power. This offers a promising and adaptable approach for feature extraction in BCI and broader neuroimaging applications.

Minimum-volume nonnegative matrix factorization (min-vol NMF) has been used successfully in many applications, such as hyperspectral imaging, chemical kinetics, spectroscopy, topic modeling, and audio source separation. However, its robustness to noise has been a long-standing open problem. In this paper, we prove that min-vol NMF identifies the groundtruth factors in the presence of noise under a condition referred to as the expanded sufficiently scattered condition which requires the data points to be sufficiently well scattered in the latent simplex generated by the basis vectors.

This paper investigates the dynamic properties of planar slider-pusher systems as a motion primitive in manipulation tasks. To that end, we construct a differential kinematic model deriving from the limit surface approach under the quasi-static assumption and with negligible contact friction. The quasi-static model applies to generic slider shapes and circular pusher geometries, enabling a differential kinematic representation of the system. From this model, we analyze differential flatness - a property advantageous for control synthesis and planning - and find that slider-pusher systems with polygon sliders and circular pushers exhibit flatness with the centre of mass as a flat output. Leveraging this property, we propose two control strategies for trajectory tracking: a cascaded quasi-static feedback strategy and a dynamic feedback linearization approach. We validate these strategies through closed-loop simulations incorporating perturbed models and input noise, as well as experimental results using a physical setup with a finger-like pusher and vision-based state detection. The real-world experiments confirm the applicability of the simulation gains, highlighting the potential of the proposed methods for