Search Results (215)

Artificial intelligence is increasingly being investigated for robot motion generation, while conventional methods remain effective for deterministic waypoint tasks. This study evaluates reinforcement learning as a state-conditioned joint-reference-generation layer at runtime, not as a replacement for classical control. A hybrid soft actor–critic (SAC)–linear-quadratic regulator (LQR) framework was implemented for a four-phase UR3e waypoint task. The SAC policy generated bounded joint-reference increments every 50 ms, while six per-joint LQR loops tracked them every 5 ms. Across 50 randomised MATLAB/Simulink episodes, the framework achieved 96% full-task success upon first entry into a strict 5 mm place region, with a first-entry distance of $4.221\pm 0.381$ mm. Frozen-policy tests under broader initial configurations, unseen waypoints, model-parameter perturbations, SAC observation-vector noise, command delay, and external torque achieved 82–100% full-task success. Under the nominal protocol, matched offline-trajectory LQR and proportional–integral–derivative (PID) baselines and a conventional inverse-kinematics (IK)–trajectory–LQR baseline each achieved 100% task completion. Post-entry continuation showed that threshold-entry success did not yield sustained sub-5 mm placement over 0.5, 1.0, and 2.0 s dwell intervals. A separate MATLAB–Robot Operating System 2 (ROS 2) fake-hardware experiment characterised execution and communication latency, with a mean round-trip latency of 5.45 ms. It excluded the Simulink plant and LQR inner loops and, therefore, represents a pre-deployment controller-pipeline evaluation rather than validation of the complete architecture on a physical robot. Full article

Long-distance free-surface water-conveyance tunnels require a spatially continuous representation of transient water depth, yet in practice flow is monitored at only a few sections. This study develops a physics-informed neural network (PINN) that reconstructs the transient water-depth field of unsteady free-surface flow in such tunnels from two- or three-point sensors. The one-dimensional Saint-Venant equations, closed with a Darcy–Weisbach steady friction term in hydraulic-radius form, are embedded as a soft constraint in the training loss, so that sparse depth observations are combined with the governing conservation laws to recover the field at unobserved interior and downstream locations. High-resolution finite-volume (FVM) solutions of a 500 m circular tunnel under a flood-rise scenario provide the reference data. The PINN reduces the relative L₂ error at unobserved sections to approximately one-quarter of that of an otherwise identical, physics-free ANN (2.50% versus 10.05% at an interior section; 4.77% versus 13.69% at an extrapolation section). Runs repeated with different random seeds confirm statistical stability at zero noise, while revealing that a minority of trainings at 10% noise converge to spurious solutions. Sensor-placement experiments, including layouts anchored at the true domain boundaries (x = 0 and 500 m), show that boundary anchoring—particularly of the upstream boundary—governs both accuracy and noise robustness: boundary-anchored two-sensor layouts remain accurate in most runs under 10–20% observation noise, whereas interior-only layouts degrade sharply. The method is presented as an offline reconstruction tool; its extension to streaming data assimilation is discussed as future work. Full article

UAVs are being employed for inspection, mapping, emergency response, and low-altitude activities. However, propeller fractures, edge wear, tip loss, imbalance, motor-bearing deterioration, eccentric shafts, and operating circumstances might damage their propulsion systems. Acoustic health monitoring is promising for this task because propulsion failures may change pressure fluctuations, blade-passing tones, broadband aerodynamic noise, and vibration-radiated sound without structural changes. Wind, reverberation, interference from surrounding rotors, speed variations, sensor-direction effects, limited datasets, and the gap between laboratory precision and field robustness in operational situations prevent acoustic UAV diagnostics from being widely used. Acoustic diagnostics of small UAV propulsion systems are the focus of this article. Acoustic emission monitoring, microphone-array and beamforming, wavelet and time–frequency feature extraction, normalized feature design, machine learning, lightweight deep learning, and noise-resilient deployment evaluation criteria are included. The reviewed studies reported promising results, including on-board microphone-array isolation of damaged rotors, acoustic-camera propeller-tip identification, and deep learning on propeller-anomaly datasets. Standardization of sensor locations, publicly accessible multi-condition acoustic datasets, cross-UAV validation, systematic selection of wavelet bases and decomposition levels, and clear computational cost reporting are still lacking. A deployment-oriented roadmap emphasizes multi-condition benchmarking, sensor placement optimization, wavelet-based noise robustness, hybrid acoustic-vibration validation, domain adaptation, uncertainty-aware decision-making, and edge AI for real-time UAV propulsion health monitoring. Full article

Conventional image acquisition requires an electrically powered image sensor to be placed directly behind the camera lens, constraining camera placement. To overcome this issue, we introduce M3-RGB as an incoherent-light fiber imaging system in which a multicore, multimode optical fiber passively relays lens images to a remotely located image sensor. Unlike conventional approaches, M3-RGB is designed to operate directly on incoherent light and requires no electrical power or active components at the sensing interface. Because propagation through the fiber yields spatially scrambled patterns, a neural network is used to reconstruct the original scene by exploiting the spatial locality preserved by the multicore structure. In a controlled optical bench setup, where a liquid crystal display monitor displays road-scene images, we construct a paired dataset of scrambled and ground-truth images and quantitatively evaluate reconstruction performance across different fiber core counts, fiber lengths, and calibration settings, utilizing the peak signal-to-noise ratio and structural similarity index measure as performance metrics. By decoupling imaging electronics from the sensing point, this passive remote image relay approach may expand sensor placement options for potential applications such as all-around perception for mobile robots and autonomous vehicles, surveillance, and inspection in confined spaces. Evaluations in real outdoor environments constitute future work. Full article

Attention-deficit/hyperactivity disorder (ADHD) involves difficulties in sustaining attention and resisting distraction. This has motivated the development of feedback-driven environments for cognitive control training. Integrating electroencephalography (EEG) sensors into Virtual Reality (VR) serious games for cognitive therapy remains relatively underexplored and requires reliable, non-invasive brain–computer interfaces. The existing solutions use multi-channel systems that primarily suffer from requiring complex hardware, while not combining motor imagery (MI) with concentration levels. Therefore, this case study evaluates the feasibility and data quality of a lightweight, cost-effective sensor configuration for real-time control of mental state. A non-invasive, eight-channel OpenBCI Cyton board was integrated with an EEG cap using the international 10–20 placement system, alongside a Meta Quest 2 headset, to capture MI and concentration signals directly from the user’s scalp. Signal acquisition was hindered by high impedance and channel railing, which required conductive gel mitigation, while mechanical tension from the VR headset strap introduced motion artifacts and noise. Nevertheless, under stable signal conditions, the optimized eight-channel sensor setup achieved a subject-specific online classification accuracy of up to 90% using the deep learning model “EEGNet”. The findings demonstrate the technical feasibility of acquiring and classifying EEG activity using a low-cost eight-channel sensor configuration in an interactive VR-BCI Serious Gaming application, provided that skin–electrode impedance and mechanical sensor interferences are managed. The results provide a basis for future investigation of such systems in cognitive-training applications, while further studies, including clinical evaluations, are required to assess their applicability in therapeutic contexts. Full article

This paper presents a novel master–slave stochastic optimization framework for the optimal siting and sizing of fixed-step capacitor banks in medium-voltage distribution networks, explicitly addressing the inherent variability of load demand that is typically neglected in conventional deterministic approaches. The proposed methodology integrates a scenario-based stochastic optimization model with a Chu and Beasley genetic algorithm (CBGA) as the master stage, which handles discrete placement decisions, and a successive-approximation power flow method (SAPF) as the slave stage, which evaluates the technical and economic performance of each candidate solution under multiple load scenarios. To capture demand uncertainties, 365 daily load realizations are generated using independent Gaussian noise with a relative standard deviation of 10% applied to each load point. These are subsequently reduced to ten representative scenarios via k-means clustering, reducing the number of power-flow evaluations per candidate solution from 365 to 10 (a 36.5-fold reduction); the reduced scenarios exhibit a low mean absolute error (MAE: <2%) with respect to the original mean, indicating faithful representation of the average load behavior, while the silhouette score is modest (approximately 0.25), consistent with the unimodal nature of the generated data and implying that the clusters are not well separated. Extensive simulations on a 33-bus test feeder considering three energy-cost-escalation scenarios (0%, 10%, and 20%) over a 20-year planning horizon demonstrate that both the deterministic and stochastic approaches reduce the total net present cost by 16.52% to 17.34% compared to the uncompensated network; the stochastic approach consistently delivers solutions that are either superior or comparable to deterministic planning (yielding up to approximately 0.16% additional cost reduction) while offering enhanced robustness against load variability. The stochastic framework offers distinct advantages, including robust solutions across a wide range of operating conditions, an inherent ability to adjust investment levels in response to probabilistic load distributions, and the ability to quantify uncertainty in decision making, with the most significant benefits observed when energy costs are low and load variability is high. The convergence of both approaches at a 20% escalation level further validates the reliability of high-resolution deterministic modeling when economic factors strongly dominate the optimization objective. This study underscores the importance of probabilistic modeling for modern distribution network planning, providing a practical and computationally efficient decision-support tool for utility planners to enhance grid resilience and operational efficiency in the context of increasing demand variability and renewable energy integration. Full article

This work presents a modular, sensor-aware, end-to-end simulation framework for optical camera communication (OCC). The framework combines modified Monte Carlo ray tracing with pixel-level camera modeling, including optical blur, shutter timing, noise, digitization, and modulation-aware signal processing. It produces physically consistent synthetic images from simulated optical propagation and enables communication performance to be estimated through image-domain signal-to-noise ratio (SNR) and theoretical bit-error-rate (BER) calculations. The simulator is applied to intra-satellite optical links as a representative case study involving confined three-dimensional geometries, line-of-sight (LOS) visibility, partial occlusion, and rolling-shutter image formation. Experimental validation under LOS conditions shows good agreement in the dominant spatial-temporal characteristics of rolling-shutter imagery, with a structural similarity index measure (SSIM) of approximately $0.80$. Simulated SNR values range from approximately $20.5$ to $22.6$ dB, compared with measured values between $20.8$ and $24.4$ dB. No bit errors are observed in the experimental sequences, corresponding to finite-length BER upper confidence bounds, while the BER values derived from the simulated images are theoretical estimates obtained from the image-domain SNR under ideal receiver assumptions. Additional simulations using a detailed 12U CubeSat model demonstrate the capability to assess emitter–receiver placement and partial geometric occlusion, including cases in which the visible portion of the source remains sufficient for bitstream decoding. By jointly modeling optical propagation, camera acquisition, and communication metrics, the proposed framework supports early-stage OCC system analysis and configuration trade-offs without requiring immediate hardware implementation. The approach is applicable to other OCC scenarios in which spatial image formation and sensor dynamics influence communication performance. Full article

Wireless sensing networks increasingly extend into obstacle-prone deployments, where physical blockage degrades reliability and open propagation exposes transmission activity. Intelligent reflecting surfaces (IRSs) establish programmable paths around obstacles while passive elements remain constrained by severe cascaded attenuation. To address the tradeoff between reliability and covertness, we propose an unmanned aerial vehicle (UAV) -assisted active-IRS architecture under probabilistic line-of-sight and non-line-of-sight propagation conditions that accounts for direct leakage from the transmitter to the warden together with residual jammer cancellation and always-on IRS circuit noise under a finite output power budget. Furthermore, bidirectional Kullback–Leibler analysis identifies the reverse divergence as the tighter restriction and converts the covertness requirement into conservative gain bounds under warden location uncertainty and relative phase uncertainty conditions between the direct and aggregate reflected fields. Subsequently, closed-form phase control for calibrated equal-gain elements and gain monotonicity reduce the joint design to an exhaustive search over the prescribed placement grid. The numerical results demonstrate a SINR advantage over passive reflection and single-element relaying across the evaluated settings. The finite-array and hardware analyses show that gain back-off enforces a prescribed covert-outage limit while direct leakage and residual self-interference remain explicitly controlled. Overall, the framework provides a transparent basis for reliable covert sensing through UAV-assisted active reflection. Full article

Wake-induced magnetic fields provide a potential non-acoustic signature for underwater-vehicle sensing, but their weak amplitudes can be masked by ocean-wave magnetic backgrounds. This study evaluates their engineering observability under representative wind–wave conditions. The wake-induced field at fixed observation points was calculated from CFD-derived wake velocities of an engineering-scale fully appended SUBOFF model using discrete Biot–Savart summation. The ocean-wave background was computed using a JONSWAP spectrum and linear wave theory, and a peak-to-background-rms SNR was used as the observability indicator. Results show that speed and diving depth strongly control the target signal. At the baseline point, increasing speed from 10 to 40 kn raised $B_{wake,max}$ from 0.0406 to 1.65 nT and SNR from −1.01 to 31.2 dB under W2. Increasing diving depth from $2D$ to $4D$ reduced $B_{wake,max}$ from 0.129 to 0.0204 nT and SNR from 9.03 to −6.99 dB. Wind speed dominated the wave background: at $U_{10}=10$ m/s, $B_{wave,rms}$ reached 0.542 nT and the Case 2 SNR decreased to −12.5 dB. Sensor placement affected both signal and background; deeper underwater sensors improved observability, whereas aerial observations suffered from weak wake-signal amplitudes. Wake-field observability is therefore jointly governed by wake source strength, ocean-wave magnetic background, and observation geometry. Full article

When a wearable gait predictor is personalized to a new user, a short calibration session may contain missing channels, sensor-placement changes, electrode shift, noise, fatigue, or other non-ideal effects. Treating all calibration samples as equally trustworthy can overpersonalize the prediction head and induce harmful drift away from the source model. We study calibration quality as a control signal for new-user personalization and propose Dynamic Trust-Region Head Adaptation (DTR-HA), a head-only adaptation rule that combines reliability-weighted calibration loss with a reliability-scaled source-head anchor. The temporal encoder is frozen, and lower reliability strengthens a soft source-head penalty. The Bilateral Lower-Limb Neuromechanical Signals dataset (BLISS) defines the target early gait-phase prediction task with 21-subject leave-one-subject-out evaluation; complete-bout calibration/test separation; and 0, 100, and 200 ms horizons. K2MUSE supplies real non-ideal calibration conditions for mechanism-level stress testing. In K2MUSE 75% bad-calibration tests with a held-out blind context-based reliability scorer, DTR-HA reduced empirical risk by 20–23% relative to plain head adaptation, improved macro-F1 by 0.018–0.022, and achieved 25/28 paired subject–condition wins with fewer negative-adaptation cases. In a secondary BLISS calibration-pollution check, DTR-HA kept macro-F1 within 0.003 of plain head adaptation while reducing calibration-induced head drift across all horizons. These results support calibration quality as an adaptation-stage control signal for stable, lightweight wearable gait personalization under unreliable calibration. Full article

We develop a versatile description of Fabry–Pérot resonators comprising internal dielectric structures, based on transfer matrices and interpreted from a non-Hermitian one-pole self-energy viewpoint. For a two-mirror cavity containing an internal slab of arbitrary thickness and refractive index, we derive closed-form expressions for transmission and identify cavity poles as zeros of the reduced denominator in the complex-frequency plane. For a weak-reflector, we obtain leading-order expressions for the pole shifts of individual modes and show that the resulting mode pulling and linewidth change are, respectively, governed by the imaginary and real parts of a single complex quantity formed by the coherent sum of two mirror-side scattering paths. These expressions provide placement criteria for dispersive or dissipative operation and support practical workflows for extracting weak-reflector parameters from measured resonance traces, predicting slab-modified cavity spectra from the empty cavity calibration, and designing doubly resonant cavities with fine constraints on slab position. We extend the consideration to multiple reflectors, with emphasis on two-membrane and three-mirror geometries, relevant to coupled filter cavities and membrane-in-the-middle architectures. The coupled-pole parametrization relates these configurations, providing a compact framework for the analysis and design of composite Fabry–Pérot elements for precision filtering and quantum-noise shaping in advanced interferometric experiments. Full article

This article investigates sensor placement strategies for 3-D time-of-arrival (TOA)-based target localization in underwater acoustic sensor networks (UASNs). To prevent the overestimation of localization performance common in idealized marine models, we derive an exact acoustic propagation time and estimate the TOA measurement variance using a non-linear ray acoustic model. Leveraging this formulation, we establish a realistic 3-D TOA measurement model that incorporates depth-dependent sound speed profiles (SSP) and spatially heterogeneous noise, where the trace of the Cramér-Rao lower bound (CRLB) serves as the optimization criterion. To solve the resulting non-convex optimization problem, we propose a MinMax k-Means algorithm to determine the optimal sensor configuration by minimizing the average of the trace of CRLB. Extensive numerical simulations demonstrate that the proposed placement strategy significantly enhances localization accuracy and robustness compared to conventional benchmarks, proving its effectiveness in complex underwater environments. Full article

Surface electromyography (sEMG) is widely used to evaluate neuromuscular activity; however, objective and reproducible methodologies for assessing signal quality in embedded acquisition systems remain limited. This study presents a multichannel embedded sEMG acquisition platform based on an ADS1298 analog front-end (Texas Instruments, Dallas, TX, USA) and an STM32H743ZIT6 microcontroller (STMicroelectronics, Geneva, Switzerland), together with a signal-quality evaluation methodology guided by signal-integrity principles derived from IEC 60601-2-40. sEMG signals were acquired at 2 kHz from 10 healthy participants during standardized submaximal isometric and controlled isotonic contractions using consistent electrode placement and acquisition procedures. Signal quality was quantified using complementary temporal and spectral metrics, including signal-to-noise ratio (SNR), power-line interference ratio (PLI), baseline drift, root mean square stability, median frequency, spectral entropy, skewness, and kurtosis. Signal conditioning reduced the baseline drift from 0.32 ± 0.23 to 0.004 ± 0.003 and reduced PLI from 0.097 ± 0.135 to 0.010 ± 0.003 while preserving contraction-dependent temporal and spectral characteristics. In addition, envelope-based Spearman correlation analysis revealed contraction-dependent intermuscular coordination patterns between the long and short heads of the biceps brachii under controlled acquisition conditions. These findings demonstrate that combining standardized acquisition protocols, objective signal quality metrics, and interpretable correlation-based analysis provides a reproducible engineering framework for evaluating embedded sEMG systems and establishes a foundation for future machine learning approaches based on larger and well-characterized physiological datasets. Full article

Effective condition monitoring (CM) in the mining industry is crucial for operational excellence, given the harsh environments, continuous operation, and high-value nature of assets. Traditional fault diagnosis methods like vibration analysis often prove inadequate due to signal noise, logistical challenges for sensor placement, and limitations in detecting subtle failures. This paper addresses these challenges by proposing an advanced contactless diagnostic system that integrates Infrared Thermography (IRT) with an optimized Convolutional Neural Network (CNN) for detecting machinery faults in mining operations. The core of the approach involves a customized ResNet-50 architecture, chosen for its inherent ability to extract hierarchical features directly from raw thermal image data, thereby circumventing the laborious and error-prone process of manual feature engineering. Recognizing the profound impact of hyperparameters on model performance, a novel optimization strategy is developed. This strategy utilizes an amended Salp Swarm Algorithm (SSA), which incorporates a Levy flight mutation strategy and improved position update mechanisms to enhance its exploration capabilities and prevent premature convergence, ensuring a thorough search of the complex hyperparameter space. The proposed methodology is rigorously evaluated using thermal images acquired from a heavy-duty belt conveyor system at the JARO S.A. mine. The optimized ResNet-50 model achieved a remarkable validation accuracy of 97.22%, demonstrating superior performance. Comparative analysis showed that our model significantly outperformed other state-of-the-art deep learning architectures, such as InceptionV3 and ResNet-18, as well as other metaheuristic optimization algorithms, yielding a 15.6% improvement over the basic SSA. This robust performance, combined with efficient convergence, underscores the model’s capacity for accurate and timely fault identification, paving the way for proactive maintenance, reduced downtime, and enhanced safety in demanding mining environments. Full article

In LoRa uplink communication, direct edge-node-to-gateway transmission is affected by path loss, thermal noise, small-scale fading, and intra-spreading-factor (intra-SF) and inter-spreading-factor (inter-SF) interference under random traffic activation. These factors reduce the signal-to-interference-plus-noise ratio (SINR) and degrade coverage reliability. To address these issues, this study proposes an integrated framework for coverage probability analysis and relay placement optimization in two-hop LoRa networks with random traffic activation. First, a two-hop LoRa uplink network model is established, consisting of edge source nodes, a decode-and-forward (DF) relay, a central gateway, and potential interfering nodes. By incorporating distance-dependent path loss, receiver-side thermal noise power, and small-scale fading gains, a unified received-power model is formulated for the desired and interfering links. Second, a Bernoulli traffic-activation indicator is assigned to each potential interfering node to characterize its random transmission state and link the traffic activation probability, the active-interferer set, and the expected number of active interferers. An interference model is then developed to quantify the effect of random traffic activation on aggregate interference over communication links. To account for multi-spreading-factor (multi-SF) coexistence in LoRa, intra-SF interference and residual inter-SF interference are incorporated into the link-level SINR criterion, and the corresponding single-hop coverage probability is derived. Finally, based on the DF relaying protocol, successful end-to-end transmission is modeled as the joint event of successful first-hop decoding and successful second-hop forwarding, and a two-hop coverage probability model is constructed. A system-level coverage probability model is also developed to capture the complementarity between direct transmission and two-hop relaying. A relay placement optimization problem is then formulated to maximize the weighted average system coverage probability across multiple traffic states. The performance of fixed, random, and optimized relay placement schemes is compared. Simulation results demonstrate that the proposed random-traffic-aware relay placement optimization method substantially improves the end-to-end coverage probability compared with fixed and random relay placement schemes, thereby enhancing the communication reliability of edge-node transmissions. Full article