Search Results (559)

The position of the center of mass directly affects the attitude control and flight safety of moving bodies such as unmanned aerial vehicles (UAVs). Therefore, high-precision measurement of the center of mass is required. Existing methods require changing the posture of the measured object multiple times. This introduces repeated positioning errors and suffers from poor equipment versatility. To address these issues, this paper proposes a static and dynamic combined measurement method for the center of mass based on multi-view vision. First, the relationship between the swing period and the pendulum length under the simple pendulum principle is analyzed. The basic principle of determining the direction of the center of mass using the line of gravity is also examined. Second, an under-constrained compound pendulum fixture is designed. A binocular vision system is used to track circular markers, perform FFT-based period verification, and fit the gravity line using singular value decomposition (SVD). Third, using a standard cubic iron block as the test object, the influence of pendulum length and swing angle on measurement accuracy is studied. Finally, experiments verify that the proposed method can obtain three-dimensional coordinates of the center of mass under a single suspension condition. The results show that with a pendulum length of 330 mm and an initial swing angle of 4°, the root mean square error of the center of mass measurement is 0.70 mm, and the maximum deviation over five repeated measurements is 1.45 mm. This method does not require repeated lifting or changes in posture. It can meet the need for in-situ, high-precision center of mass measurement of UAVs and other aircraft. Full article

Artificial Intelligence (AI) and Machine Learning (ML) are increasingly driving the digital transformation of manufacturing systems, enabling the transition from conventional process operation toward intelligent, adaptive, and data-centric production environments. This work presents the development of AI-enabled advisory systems for casting processes, integrating singular value decomposition (SVD)-based reduced-order models with a Variational Autoencoder with Arbitrary Conditioning (AC-VAE) and hybrid simulation frameworks to support real-time process prediction and optimization. The proposed approach leverages manufacturing data to establish predictive models capable of rapidly evaluating process conditions, optimizing operating parameters, and enhancing product quality while reducing material waste, energy consumption, and production costs. By combining physics-based understanding with AI-driven analytics, the framework facilitates real-time decision support, adaptive process control, and continuous performance improvement within modern manufacturing ecosystems. These capabilities contribute to the broader objectives of Industry 4.0 and emerging Industry 5.0 paradigms, including automation, connectivity, operational resilience, sustainability, and human-centered manufacturing. A representative Horizontal Direct Chill (HDC) continuous casting case study is presented to demonstrate the practical implementation of the framework, encompassing database generation, model training, validation, and deployment of predictive advisory tools for real-time manufacturing applications. Full article

Audio magnetotelluric (AMT) sounding is an essential tool for mineral exploration and subsurface electrical structure imaging. However, persistent cultural noise can severely degrade AMT data quality, particularly in mining areas with strong anthropogenic interference. In this study, we introduce an improved K-Singular Value Decomposition (K-SVD) dictionary learning method for time-domain AMT data denoising, incorporating three key innovations: (1) automatic identification of noise-contaminated segments using local kurtosis; (2) adaptive dictionary initialization combining principal component analysis with Gaussian perturbation, which accelerates convergence and avoids local minima; and (3) online atom screening and updating to maintain the noise specificity of the learned atoms. The proposed method learns noise morphology directly from raw AMT data without requiring reference stations or external training datasets. Synthetic experiments demonstrate that the method more effectively suppresses square-wave and charge-discharge noise compared with wavelet thresholding methods and conventional K-SVD, while better preserving signal morphology. Field applications further show that it significantly enhances the quality of observed data and apparent resistivity-phase curves compared with the widely used robust estimation method. The proposed method provides an efficient, reference-free solution for AMT data denoising in mining areas with strong cultural noise. Full article

Under moderate-to-high wind conditions, low-frequency wind direction modulation signals in X-band marine radar images are strongly coupled with wave textures, sea clutter, and blind-zone interference, which degrades wind direction retrieval accuracy. To address this problem, this study proposes a wind direction retrieval method based on two-dimensional dual-tree complex wavelet transform (2D-DTCWT), convolutional sparse coding (CSC), and maximum-energy radial rings. First, 2D-DTCWT is used to suppress wave textures and local noise in the wavelet domain while enhancing low-frequency wind direction modulation signals. Then, K–singular value decomposition (K-SVD) learns the energy distribution characteristics of wind signals, and CSC obtains the spatial response distribution of wind energy in radar images. Finally, the maximum-energy radial ring is adaptively identified, and azimuthal energy statistics within this ring are fitted using a cosine-squared function. The proposed method was evaluated using X-band marine radar data collected during sea trials in the coastal waters of Zhejiang, China. On the 900-sample main validation dataset, the proposed method achieved the highest correlation coefficient (CC) of 0.85 and an overall root mean square error (RMSE) of 4.24°, reducing the RMSE by 43.0% and 66.7% compared with conventional single-curve fitting and extended-bow-heading DWT, respectively. The results demonstrate improved robustness under both upwind and downwind blind-zone conditions. Full article

In the preliminary vehicle design process, Finite Element Analysis (FEA) and Multi-Body Dynamics (MBD) simulations require detailed physical parameters that are not available during the conceptual design phase. This study proposes a novel analytical algorithm that uses a virtual weighting approach to rapidly establish a reasonable baseline for these parameters, serving as an efficient analytical precursor to the physical tests, complex dynamic simulations, and optimization methods conventionally applied in later stages. In this study, a dual-layer model for rapid geometric assessment of stability indices and vertical loads in hyperstatic 8 × 8 vehicles is proposed. The model consists of a prognostic Weighted Singular Value Decomposition (SVD) layer and an operative Weighted Pseudo-Inverse (WPI) layer. In the SVD layer, a spectral mode alignment technique is proposed to evaluate the load transmission capacity to predict the stability limits under worst-case operating conditions including extreme maneuvers and wheel failures. In the WPI layer, the optimal distribution of wheel loads is computed under different operating conditions. For a uniform vehicle configuration, a high-resolution continuous sweep of lateral acceleration identifies the exact wheel lift-off point at 0.8621 g, perfectly aligning with the theoretical Static Stability Factor (SSF). By employing a virtual weighting strategy instead of relying on traditional exhaustive physical parameters, this parameter-independent framework provides an analytical load-boundary evaluation and determines the theoretical topological capacity, thereby acting as an essential tool for preliminary conceptual design prior to detailed MBD and FEA analyses. Full article

Remote sensing images are frequently degraded by occlusions and missing observations, which significantly affect subsequent interpretation and analysis. Matrix completion provides an effective solution for recovering incomplete data; however, existing deep learning-based approaches often rely on random initialization, resulting in slow optimization and limited reconstruction quality under severe missing conditions. To address these issues, this paper proposes a two-stage neural network-based matrix completion framework that combines SVD-guided low-rank modeling with convolutional feature learning. Specifically, truncated singular value decomposition (SVD) is first employed to initialize the network and provide a coarse reconstruction by jointly modeling the global low-rank structure and nonlinear image representations. A U-Net-based convolutional autoencoder is then used to refine the reconstruction by exploiting local spatial correlations and multi-scale features. In addition, a channel aggregation strategy is introduced to improve structural consistency for multi-channel remote sensing images. The proposed framework adopts a training-data-free optimization paradigm, eliminating the need for external training datasets by optimizing the network parameters directly for each input image. Experimental results on synthetic and real remote sensing images demonstrate that the proposed method consistently outperforms conventional matrix completion methods and achieves competitive performance compared with recent deep learning approaches, particularly under random missing patterns and high missing-rate scenarios. Full article

Conventional linear isolators struggle to combine high static load-bearing capacity with effective low-frequency vibration isolation. To address this limitation, this study proposes an inclined rhombic crab-inspired quasi-zero-stiffness (I-QZS) isolator. A generalized static model is established based on the segmented linkage of crab walking legs. A physics-constrained NSGA-II algorithm is used to optimize the key geometric parameters while preventing bistable snap-through by imposing a positive-stiffness constraint over the full stroke. A stiffness-compensation strategy bridges the gap between the ideal rigid-body model and the actual 3D-printed compliant structure. The dynamic response is represented by a cubic polynomial restoring-force model, and the resulting equations are solved using the incremental harmonic balance method with SVD (singular value decomposition)-based null-space continuation. Large-amplitude excitation experiments show that the I-QZS shifts the resonance peak to 0.77 Hz, reducing the peak frequency by 64.19% and the peak transmissibility by 67.06% relative to a linear isolator, while substantially broadening the isolation bandwidth. Bifurcation analysis further identifies stability boundaries for engineering design. These results provide an integrated theoretical and experimental framework for ultra-low-frequency passive vibration isolation. Full article

The propagation characteristics of ultrasonic guided waves (UGWs) in rails, together with the inherently complex inspection environment for in-service rails, pose a significant challenge for the effective detection of rail fractures or damage using UGW. To address these issues, this work proposes a feature extraction method for defect-related UGW signals in the structural health monitoring of rails. The proposed approach integrates singular value decomposition (SVD), principal component analysis (PCA), and independent component analysis (ICA). First, to eliminate periodic interference, the constructed data matrix is processed using a Hankel matrix combined with SVD, and the analysis matrix is then reconstructed for subsequent ICA processing. Second, to reduce computational complexity, the dimensionality of the analysis matrix is reduced using PCA. Finally, the independent components and their corresponding weight vectors are obtained by applying ICA to the analysis matrix. By examining whether the resulting weight vectors exhibit significant step characteristics, it is possible to determine whether the rail is defective. Experimental validation demonstrates that the proposed feature extraction method is both reliable and effective. Full article

Alzheimer’s disease (AD) is the most prevalent neurodegenerative disorder worldwide, affecting approximately 56.9 million people in 2021 and projected to reach 152 million by 2050. Its defining pathological features, amyloid-beta plaques and neurofibrillary tangles, accumulate for up to two decades before cognitive symptoms emerge, placing the preclinical and mild cognitive impairment (MCI) stages at the centre of the early detection problem. Despite this, current diagnostic practice in routine clinical settings remains unreliable, with post-mortem studies placing the specificity of clinical AD diagnosis between 44.3 and 70.8% even in specialist memory clinics. Machine learning (ML) and deep learning (DL) applied to neuroimaging and electrophysiological data have emerged as candidate tools for closing this diagnostic gap, yet whether the accuracy figures reported in published studies translate into clinically useful performance on independent data remains unresolved. This study presents a structured comparative review of machine learning and deep learning methods reported across four publicly available Alzheimer’s disease datasets, namely the Alzheimer’s Disease Neuroimaging Initiative (ADNI), the Open Access Series of Imaging Studies (OASIS), the OpenNeuro ds004504 electroencephalography (EEG) dataset, and the Kaggle Alzheimer’s magnetic resonance imaging (MRI) dataset. Thirteen model families are examined through the published literature rather than through new experiments, and for each model and dataset combination, the best accuracy reported in the source study is recorded alongside the model’s mathematical formulation. All performance figures reported in this abstract and throughout the paper are taken from the published studies reviewed, not from new experiments conducted by the authors. Across the reviewed studies, deep learning architectures pre-trained on ImageNet and fine-tuned on neuroimaging data are reported to produce the highest accuracy on MRI classification tasks. Residual Network (ResNet)-101 is reported at 98.21 percent on ADNI and 97.45 percent on OASIS, while the IncepRes fusion architecture reaches 98.35% on OASIS by combining multi-scale feature extraction from InceptionV3 with residual connectivity from ResNet152V2. Traditional machine learning classifiers remain competitive on tabular clinical and biomarker data, with Extreme Gradient Boosting (XGBoost) reaching 91% on ADNI multiclass features. Logistic Regression achieves 82 to 85% on binary ADNI classification and is the only classifier in this review that provides explicit per-feature prediction contributions without post hoc tooling. Gaussian Naïve Bayes achieves 80 to 83% on the same task. On the OpenNeuro EEG dataset, K-nearest neighbours (KNN) with singular value decomposition (SVD) entropy features achieves 91% binary accuracy, with feature engineering quality determining performance more reliably than classifier architecture. Eight principal findings emerge from the cross-dataset analysis. Binary classification consistently outperforms multiclass by 10 to 30% across all datasets, reflecting the genuine biological ambiguity of the mild cognitive impairment category. Dataset size and augmentation predict reported accuracy more reliably than model architecture. Ensemble methods outperform individual classifiers by 5 to 8% in nearly every imaging study. Deeper architectures can overfit small clinical cohorts. EEG models trail MRI models by approximately 10 to 15% on comparable binary tasks. Cross-dataset generalisation has not been systematically evaluated in most studies, and the few that have tested it report accuracy drops of 5 to 10% or more when models encounter data from different scanners or cohorts. Eight recurring limitations constrain the clinical utility of these findings. Small sample sizes and limited demographic diversity, severe class imbalance inflating raw accuracy metrics, poor cross-dataset generalisation driven by scanner heterogeneity, limited deep learning interpretability, the dominance of binary over multiclass tasks, the absence of longitudinal modelling despite available datasets, inadequate standardisation of preprocessing and evaluation protocols, and the signal-to-noise ratio constraints specific to EEG recordings of elderly patients collectively define the gap between benchmark performance and clinical readiness. Future work must prioritise multi-centre training cohorts, multimodal fusion architectures, longitudinal progression modelling, and standardised interpretability evaluation as non-optional requirements for any system intended for clinical deployment. Full article

Sixth-generation (6G) networks require advanced non-orthogonal multiple access (NOMA) schemes to support the technical requirements of dense massive machine-type communications (mMTC). While sparse-code multiple access (SCMA) improves uplink spectral efficiency, its operation within the complex two-dimensional domain ($\mathbb{C}$) generates spatial congestion and high error rates under high-load scenarios. Furthermore, when using dual-polarization transceivers, cross-polarization discrimination leakage is not efficiently exploited because it operates in a conventional 2D environment. This work proposes a hypercomplex transmission architecture, called quaternionic SVD-SCMA (Q-SVD-SCMA), which maps SCMA codewords to a quaternionic group ($Q_8$) in ${\mathbb{R}}^4$. The proposed scheme uses purely imaginary spatial rotators to project overlapping signals onto mutually orthogonal geometric subspaces, thus mitigating interference between users. On the receiver side, a quaternionic sphere decoder (Q-SD) is proposed to evaluate the minimum quaternionic Euclidean distance (MQED) to provide near-optimal detection. Computational simulations are performed under a doubly polarized Rayleigh fading channel with energy normalization to decouple arbitrary power-scale topological gains. To evaluate system performance, both the symbol error rate (SER) and the bit error rate (BER) are used. The results obtained demonstrate that Q-SVD-SCMA effectively transforms Cross-Polarization Discrimination (XPD) leakage into spatial diversity gain. The hypercomplex 4D architecture proposed in this work eliminates the interference-induced error threshold and limits the bit error penalty through multidimensional Gray mapping, providing a scalable and highly reliable physical layer framework for overloaded 6G scenarios, unlike its conventional 2D predecessors (C-SCMA and SVD-SCMA). Full article

Recommender systems face fundamental challenges, including extreme data sparsity and noisy rating observations. We propose L21RSVD, a robust latent factor model that employs L21 norm regularization within a Singular Value Decomposition framework to mitigate the impact of outliers while preserving low-rank structure. Unlike conventional L2-regularized approaches, our formulation induces group sparsity in the latent factor space, yielding more discriminative user and item representations. We derive three optimization variants: standard L21RSVD, L21RSVD without the squared term, and coefficient-free adaptive L21RSVD. Building upon these, we introduce a fusion strategy that adaptively aggregates predictions based on local data density. Extensive experiments on benchmark datasets demonstrate that L21RSVD substantially outperforms classical collaborative filtering and the SVD-type model. The proposed fusion model achieves state-of-the-art performance, reducing RMSE by up to 31.21% and MAE by up to 42.99% relative to baseline methods. Full article

In order to address the issue of rapidly divergent positioning errors of a single-platform inertial navigation system (INS) in GNSS-denied environments, this paper proposes a cross-platform cooperative navigation method based on relative range and angle measurements. The observability of the cooperative navigation system under different motion strategies is investigated using Fisher information matrix (FIM) right null-space analysis combined with singular value decomposition (SVD). The results show that with relative range and angle measurement constraints, all inertial sensor biases can be effectively estimated by two strapdown inertial navigation systems (SINSs) moving along a simple trajectory, thereby improving the navigation accuracy. Experimental results demonstrate that compared to the autonomous navigation mode, the average positioning accuracy of the two SINSs improves by 77.4% and 68.4% respectively after 3 h of cooperative navigation along the prescribed trajectory. Using relative range and angle measurements, the proposed method requires only two SINSs and relatively simple planar motion, without the need for high-precision reference benchmarks, complex three-dimensional excitation trajectories, or turntable modulation. It reduces system complexity and motion requirements, providing an effective and easy-to-implement solution for ground vehicular positioning and orientation and other cross-platform cooperative navigation tasks in GNSS-denied environments. Full article

Detecting weak radar targets in complex sea conditions is inherently challenging due to non-stationary sea clutter and sea spikes. Furthermore, traditional dictionary learning algorithms for clutter suppression suffer from high computational complexity. To address these issues, this paper proposes an efficient sea clutter suppression method cascading Block Coordinate Descent (BCD)-accelerated dictionary learning with Tunable Q-factor Wavelet Transform (TQWT) denoising. During dictionary learning, a BCD strategy replaces global Singular Value Decomposition (SVD) with analytical optimization. Combined with an adaptive soft-thresholding operator, this enables low-complexity joint optimization of dictionary atoms and sparse coefficients, drastically reducing training time. Subsequently, a batch-adaptive Orthogonal Matching Pursuit (OMP) algorithm featuring Gram matrix precomputation and a dual-stop mechanism achieves efficient reconstruction and preliminary cancellation of clutter components. Finally, TQWT is applied to filter out residual non-stationary clutter and noise by leveraging its narrowband feature representation and shift invariance. Experiments on measured radar data from the IPIX database and datasets published by the Journal of Radars demonstrate that the proposed method significantly outperforms traditional K-SVD-based algorithms. Specifically, it improves the average signal-to-clutter-plus-noise ratio (SCNR) by 17.48 dB and requires a total execution time of only 7.99 s, achieving a highly favorable trade-off between suppression performance and computational efficiency. Full article

To address the poor computational efficiency of traditional fixed-stratification methods in very low frequency (VLF) ionospheric propagation modeling, this paper proposes a dynamic stratification algorithm. First, filtering optimization is applied to the electron density, and dynamic adaptive stratification is implemented in the vertical direction. By establishing a nonlinear mapping relationship between the electron density gradient and the stratification thickness, the algorithm integrates dynamic ionospheric stratification with a hybrid regularization algorithm for the transmission matrix. Specifically, Singular Value Decomposition (SVD) and dynamic truncation techniques are employed to process the transmission matrix, effectively resolving the numerical ill-posedness in regions with abrupt ionospheric changes. This enables high-precision calculation of reflection coefficients in the 3–30 kHz frequency band. By tuning parameters such as the reference stratification thickness and adjustment factors, an optimized stratification model and an algorithm quality evaluation coefficient are obtained. The simulation results demonstrate that, compared with fixed stratification, the proposed algorithm achieves an average relative error of 4.7% for the reflection coefficient in the VLF range while improving computational efficiency by more than 50%. This provides a promising approach for efficient and high-precision prediction of VLF wave propagation. Full article

Missing values commonly exist in dam inflow flood discharge monitoring data, which hinders flood analysis, risk assessment and reservoir scheduling. Aiming at the problems of insufficient imputation accuracy and the difficulty in adaptive threshold selection of traditional Singular Value Decomposition (SVD) in flood discharge data with strong fluctuations and high noise, this study introduces a method for filling in missing dam inflow flood discharge based on Dam Monitoring Data Reconstruction Model (DSVD). The method constructs a non-repeating sequence monitoring matrix, introduces a hard singular value threshold for adaptive denoising, and completes time series data imputation combined with a weight optimization model, which effectively improves the imputation accuracy of strongly fluctuating flood discharge data. Taking the measured inflow flood discharge data of Jinjiaba Reservoir in Chongqing as the research object, this study systematically analyzes the influence of column-to-row ratio (R_a) and data missing rate on imputation performance, and conducts a comparative verification against other models. Experimental results indicate that the optimal R_a value is 6. The coefficient of determination (R²) stays above 0.830 within a missing rate range of 5–40%, showing strong robustness against data loss. Compared with other benchmark models, the method has the highest R² (0.875) and the lowest Root Mean Square Error (RMSE, 7.771), exhibiting stronger adaptability to mountainous flood discharge data with steep rise and fall characteristics. The research findings provide a new method for the high-precision recovery of missing dam inflow flood discharge data and reliable data support for reservoir flood risk analysis and safe operation. Full article