Search Results (492)

Traffic flow forecasting plays a significant role in intelligent transportation systems (ITSs) and is instructive for traffic planning, management and control.Increasingly complex traffic conditions pose further challenges to the traffic flow forecasting. While improving the accuracy of model forecasting, the parameter effectiveness of the model is also an issue that cannot be ignored. In addition, existing traffic prediction models have failed to organically integrate data with well-designed model architectures. Therefore, to address the above two issues, we propose the STAE-BiSSSM model as a solution. STAE-BiSSSM consists of Spatio-Temporal Adaptive Embedding (STAE) and Bidirectional Selective State Space Model (BiSSSM), where STAE aims to process features to obtain richer spatio-temporal feature representations. BiSSSM is a novel structural design serving as an alternative to Transformer, capable of extracting patterns of traffic flow changes from both the forward and backward directions of time series with much fewer parameters. Comparative tests between baseline models and STAE-BiSSSM on five real-world datasets illustrates the advance performance of STAE-BiSSSM. This is especially so on METRLA and PeMSBAY datasets, compared with the SOTA model STAEformer. In the short-term forecasting task (horizon: 15min), MAE, RMSE and MAPE of STAE-BiSSSM decrease by 1.89%/13.74%, 3.72%/16.19% and 1.46%/17.39%, respectively. In the long-term forecasting task (horizon: 60min), MAE, RMSE and MAPE of STAE-BiSSSM decrease by 3.59%/13.83%, 7.26%/16.36% and 2.16%/15.65%, respectively. Full article

Model predictive control (MPC) has emerged as a favorable control approach for PMSM drives, though its practical deployment is frequently hindered by superior computational complexity and execution burden. This paper presents four finite control set MPC (FCS-MPC) techniques applied to a two-level inverter-fed PMSM drive. Two of the approaches are conventional methods, while the other two are novel developed strategies proposed in this paper. The novel techniques focus on significantly decreasing computational burdens by employing an efficient space-vector selection mechanism that quickly selects the optimum switching vector without exhaustive evaluation. A comprehensive comparative assessment of all four control methods is conducted under various operating conditions, evaluating their dynamic and steady-state performance, computational requirements, and real-time feasibility. Simulation results demonstrate that the proposed techniques achieve a significant reduction in computational effort and faster processing, up to 39.65% faster than conventional full-state evaluation, while maintaining control performances comparable to conventional techniques. These results highlight the potential of the proposed MPC approaches to bridge the gap between advanced control theory and practical implementation in real-time PMSM drive systems, providing effective solutions for installing high-performance PMSM drives on hardware with limited resources. Full article

Facial beauty prediction (FBP) is a cutting-edge task in deep learning that aims to equip machines with the ability to assess facial attractiveness in a human-like manner. In human perception, facial beauty is strongly associated with facial symmetry, where balanced structures often reflect aesthetic appeal. Leveraging symmetry provides an interpretable prior for FBP and offers geometric constraints that enhance feature learning. However, existing multi-task FBP models still face challenges such as limited annotated data, insufficient frequency–temporal modeling, and feature conflicts from task heterogeneity. The Mamba model excels in feature extraction and long-range dependency modeling but encounters difficulties in parameter sharing and computational efficiency in multi-task settings. In contrast, mixture-of-experts (MoE) enables adaptive expert selection, reducing redundancy while enhancing task specialization. This paper proposes MoMamba, a multi-task decoder combining Mamba’s state-space modeling with MoE’s dynamic routing to improve multi-scale feature fusion and adaptability. A detail enhancement module fuses high- and low-frequency components from discrete cosine transform with temporal features from Mamba, and a state-aware MoE module incorporates low-rank expert modeling and task-specific decoding. Experiments on SCUT-FBP and SCUT-FBP5500 demonstrate superior performance in both classification and regression, particularly in symmetry-related perception modeling. Full article

Spectrum prediction is essential for cognitive radio, enabling dynamic management and enhanced utilization, particularly in multi-band environments. Yet, its complex spatiotemporal nature and non-stationarity pose significant challenges for achieving high accuracy. Motivated by this, we propose a multi-scale Mamba-based multi-band spectrum prediction method. The core Mamba module combines Bidirectional Selective State Space Models (SSMs) for long-range dependencies and dynamic convolution for local features, efficiently extracting spatiotemporal characteristics. A multi-scale pyramid and adaptive prediction head select appropriate feature levels per prediction step, avoiding full-sequence processing to ensure accuracy while reducing computational cost. Experiments on real-world datasets across multiple frequency bands demonstrate effective handling of spectrum non-stationarity. Compared to baseline models, the method reduces root mean square error (RMSE) by 14.9% (indoor) and 7.9% (outdoor) while cutting GPU memory by 17%. Full article

Modeling mobile robots is crucial to odometry estimation, control design, and navigation. Classical state-space models (SSMs) have traditionally been used for system identification, while recent advances in deep learning, such as Long Short-Term Memory (LSTM) networks, capture complex nonlinear dependencies. However, few direct comparisons exist between these paradigms. This paper compares two multivariate modeling approaches for a differential drive robot: a classical SSM and an LSTM-based recurrent neural network. Both models predict the robot’s linear (v) and angular ($\omega$) velocities using experimental data from a five-minute navigation sequence. Performance is evaluated in terms of prediction accuracy, odometry estimation, and computational efficiency, with ground-truth odometry obtained via a SLAM-based method in ROS2. Each model was tuned for fair comparison: order selection for the SSM and hyperparameter search for the LSTM. Results show that the best SSM is a second-order model, while the LSTM used seven layers, 30 neurons, and 20-sample sliding windows. The LSTM achieved a FIT of 93.10% for v and 90.95% for $\omega$, with an odometry RMSE of 1.09 m and 0.23 rad, whereas the SSM outperformed it with FIT values of 94.70% and 91.71% and lower RMSE (0.85 m, 0.17 rad). The SSM was also more resource-efficient (0.00257 ms and 1.03 bytes per step) compared to the LSTM (0.0342 ms and 20.49 bytes). The results suggest that SSMs remain a strong option for accurate odometry with low computational demand while encouraging the exploration of hybrid models to improve robustness in complex environments. At the same time, LSTM models demonstrated flexibility through hyperparameter tuning, highlighting their potential for further accuracy improvements with refined configurations. Full article

Under large-scale farming conditions, automated sow estrus detection is crucial for improving reproductive efficiency, optimizing breeding management, and reducing labor costs. Conventional estrus detection relies heavily on human expertise, a practice that introduces subjective variability and consequently diminishes both accuracy and efficiency. Failure to identify estrus promptly and pair animals effectively lowers breeding success rates and drives up overall husbandry costs. In response to the need for the automated detection of sows’ estrus states in large-scale pig farms, this study proposes a method for detecting sows’ vulvar status and estrus based on multi-dimensional feature crossing. The method adopts a dual optimization strategy: First, the Bi-directional Feature Pyramid Network—Selective Decoding Integration (BiFPN-SDI) module performs the bidirectional, weighted fusion of the backbone’s low-level texture and high-level semantic, retaining the multi-dimensional cues most relevant to vulvar morphology and producing a scale-aligned, minimally redundant feature map. Second, by embedding a Spatially Enhanced Attention Module head (SEAM-Head) channel attention mechanism into the detection head, the model further amplifies key hyperemia-related signals, while suppressing background noise, thereby enabling cooperative and more precise bounding box localization. To adapt the model for edge computing environments, Masked Generative Distillation (MGD) knowledge distillation is introduced to compress the model while maintaining the detection speed and accuracy. Based on the bounding box of the vulvar region, the aspect ratio of the target area and the red saturation features derived from a dual-threshold method in the HSV color space are used to construct a lightweight Multilayer Perceptron (MLP) classification model for estrus state determination. The network was trained on 1400 annotated samples, which were divided into training, testing, and validation sets in an 8:1:1 ratio. On-farm evaluations in commercial pig facilities show that the proposed system attains an 85% estrus detection success rate. Following lightweight optimization, inference latency fell from 24.29 ms to 18.87 ms, and the model footprint was compressed from 32.38 MB to 3.96 MB in the same machine, while maintaining a mean Average Precision (mAP) of 0.941; the accuracy penalty from model compression was kept below 1%. Moreover, the model demonstrates robust performance under complex lighting and occlusion conditions, enabling real-time processing from vulvar localization to estrus detection, and providing an efficient and reliable technical solution for automated estrus monitoring in large-scale pig farms. Full article

The increasing complexity and commercialization of the global space sector have elevated the strategic role of public–private partnerships (PPPs). However, the criteria for selecting suitable PPP models remain underexplored, particularly regarding the influence of national policy and market environments. This study proposes a decision-making framework that links six indicators—national strategic goals, government role preferences, regulatory structures, capital access, private-sector capabilities, and commercial demand—to four distinct PPP models in the space sector. Drawing on Eisenhardt’s multi-case theory-building methodology, this study analyzes PPP evolution in four countries representing mature, emerging, and nascent countries: the United States, Japan, India, and the United Arab Emirates. The cross-case analysis reveals that high-autonomy PPP models emerge only when institutional, financial, and market factors are systemically aligned. Divergence in PPP forms is driven not solely by technical capabilities but also by governance postures and regulatory designs. The findings contribute to addressing ongoing challenges related to policy reform and increasing private-sector involvement in the space sector by developing a practical decision-making tool for public and private-sector actors engaged in space governance. Specifically, the diagnostic framework enables stakeholders to assess national readiness and select appropriate PPP models. It also supports strategic planning by highlighting the reforms and capacity-building measures required for countries with nascent and emerging economies to transition from government-led missions to commercially integrated space ecosystems. Full article

Breast ultrasound imaging functions as a vital radiation-free detection tool for breast cancer, yet its low contrast, speckle noise, and interclass variability make automated interpretation difficult. In this paper, we introduce UltraScanNet as a specific deep learning backbone that addresses breast ultrasound classification needs. The proposed architecture combines a convolutional stem with learnable 2D positional embeddings, followed by a hybrid stage that unites MobileViT blocks with spatial gating and convolutional residuals and two progressively global stages that use a depth-aware composition of three components: (1) UltraScanUnit (a state-space module with selective scan gated convolutional residuals and low-rank projections), (2) ConvAttnMixers for spatial channel mixing, and (3) multi-head self-attention blocks for global reasoning. This research includes a detailed ablation study to evaluate the individual impact of each architectural component. The results demonstrate that UltraScanNet reaches 91.67% top-1 accuracy, a precision score of 0.9072, a recall score of 0.9174, and an F1-score of 0.9096 on the BUSI dataset, which make it a very competitive option among multiple state-of-the-art models, including ViT-Small (91.67%), MaxViT-Tiny (91.67%), MambaVision (91.02%), Swin-Tiny (90.38%), ConvNeXt-Tiny (89.74%), and ResNet-50 (85.90%). On top of this, the paper provides an extensive global and per-class analysis of the performance of these models, offering a comprehensive benchmark for future work. The code will be publicly available. Full article

Skeleton-based action recognition networks have widely adopted the approach of Graph Convolutional Networks (GCN) due to their superior capabilities in modeling data topology, but several key issues still require further investigation. Firstly, the graph convolutional network extracts action features by applying temporal convolution to each key point, which causes the model to ignore the temporal connections between different important points. Secondly, the local receptive field of graph convolutional networks limits their ability to capture correlations between non-adjacent joints. Motivated by the State Space Model (SSM), we propose an Action Spatio-temporal Aggregation Network, named ActionMamba. Specifically, we introduce a novel embedding module called the Action Characteristic Encoder (ACE), which enhances the coupling of temporal and spatial information in skeletal features by combining intrinsic spatio-temporal encoding with extrinsic space encoding. Additionally, we design an Action Perception Model (APM) based on Mamba and GCN. By effectively combining the excellent feature processing capabilities of GCN with the outstanding global information modeling capabilities of Mamba, APM is able to comprehend the hidden features between different joints and selectively filter information from various joints. Extensive experimental results demonstrate that ActionMamba achieves highly competitive performance on three challenging benchmark datasets: NTU-RGB+D 60, NTU-RGB+D 120, and UAV–Human. Full article

This paper investigates modulation recognition technology for high-order modulated signals. Addressing the issue that existing deep learning-based modulation recognition methods struggle to effectively capture the features of long sequence signals in high-order modulation, we propose a ConvMamba model that integrates convolutional neural networks (CNNs) with the Mamba2 architecture. By employing a selective state-space model, the ConvMamba effectively captures the temporal dependencies in long sequence signals. It also combines the local feature extraction capability of CNNs with a soft-thresholding denoising module, forming a hybrid structure that possesses both global modeling and noise resistance capabilities. The evaluation results on the Sig53 dataset, which contains a rich variety of high-order modulations, demonstrate that compared to traditional CNN- or Transformer-based architectures, ConvMamba achieves a better balance between computational efficiency and recognition accuracy. Compared to Transformer models with similar performance, ConvMamba reduces computational complexity by over 60%. Compared to CNN models with comparable computational resource consumption, ConvMamba significantly improves recognition accuracy. Therefore, ConvMamba shows a distinct advantage in processing high-order modulated signals with long sequences. Full article

This study presents an integrated machine learning framework to evaluate the aging states of lithium-ion batteries and to classify them according to their second-life application potential. The methodology combines two key components: a set of regression models to estimate critical health indicators, such as capacity and internal resistance, and a classification stage to group batteries based on these parameters. The proposed models were trained and validated using the NASA Battery Aging Datasets. Through an in-depth analysis of environmental conditions, the study identifies their influence on aging metrics, reinforcing the relevance of the input features selected. Furthermore, a clustering-based approach was employed to validate the classification performance and to reveal the link between a battery’s operation and its aging in the Euclidean space. The results show accurate predictions without signs of overfitting or underfitting, and the classification framework proved robust across the evaluated cases. This suggests that the proposed method can serve as a scalable and adaptable tool to guide battery repurposing strategies. Overall, the findings contribute to bridging the gap between battery diagnostics and real-world energy storage applications, offering practical insights to optimize second-life deployment. Full article

The blocking job shop problem (BJSSP) is a variant of the classical job shop problem, where a job completed on one machine cannot be transferred to the next machine until the latter becomes available, causing the current machine to remain blocked. Many real-world problems have been modeled as BJSSP. This problem is classified as strongly NP-hard. The existing literature indicates that some proposed methods fail to guarantee the generation of feasible solutions during the search process, necessitating solution reconstruction. In this study, we developed an Estimation Distribution Algorithm (EDA) designed to operate strictly within the BJSSP feasible solution space. The experiments showed that no specific levels of the factors were found to be significant in determining the quality of the solutions found by the EDA across all problem sets. On the other hand, incorporating assignment probabilities into the construction of EDA solutions facilitated diversity among the generated solutions. The runtime of the EDA for the proposed problem sets in the experiments, using the factor levels defined, ranged from 20 s for problems with 10 jobs and 5 machines to 600 s for those with 30 jobs and 10 machines. The proposed EDA begins by generating a random initial population of solutions. The makespan (Cmax) is then computed for each individual. The next generation is composed by three sets of individuals. The first group corresponds to a percentage of the best-performing individuals of the current generation. The second group is generated by using a probability distribution function derived from the structure of the selected solutions, while the remaining individuals are created randomly to enhance the algorithm’s exploratory capability. Our results demonstrate that the proposed method improves upon some of the previously best-known results in the current state of the art. Full article

Urban logistics face complexity due to traffic congestion, fleet heterogeneity, warehouse constraints, and driver workload balancing, especially in the Heterogeneous Multi-Trip Vehicle Routing Problem with Time Windows and Time-Varying Networks (HMTVRPTW-TVN). We develop a mixed-integer linear programming (MILP) model with dual-peak time discretization and exact linearization for heterogeneous fleet coordination. Given the NP-hard nature, we propose a Hyper-Heuristic based on Cumulative Reward Q-Learning (HHCRQL), integrating reinforcement learning with heuristic operators in a Markov Decision Process (MDP). The algorithm dynamically selects operators using a four-dimensional state space and a cumulative reward function combining timestep and fitness. Experiments show that, for small instances, HHCRQL achieves solutions within 3% of Gurobi’s optimum when customer nodes exceed 15, outperforming Large Neighborhood Search (LNS) and LNS with Simulated Annealing (LNSSA) with stable, shorter runtime. For large-scale instances, HHCRQL reduces gaps by up to 9.17% versus Iterated Local Search (ILS), 6.74% versus LNS, and 5.95% versus LNSSA, while maintaining relatively stable runtime. Real-world validation using Shanghai logistics data reduces waiting times by 35.36% and total transportation times by 24.68%, confirming HHCRQL’s effectiveness, robustness, and scalability. Full article

The Zenith Tropospheric Delay (ZTD) is a significant atmospheric error affecting the accuracy of the Global Navigation Satellite System (GNSS). Accurate estimation of the ZTD is essential for enhancing GNSS positioning precision and plays a critical role in meteorological and climate-related applications. To address the limitations of current deep learning models in capturing long-term dependencies in ZTD sequences and overcoming computational inefficiencies, this study proposes SSMB-ZTD—an efficient deep learning model based on an improved selective State Space Model (SSM) architecture. To address the challenge of modeling long-term dependencies, we introduce a joint time and position embedding mechanism, which enhances the model’s ability to learn complex temporal patterns in ZTD data. For improving efficiency, we adopt a lightweight selective SSM structure that enables linear-time modeling and fast inference for long input sequences. To assess the effectiveness of the proposed SSMB-ZTD model, this study employs high-precision Zenith Tropospheric Delay (ZTD) products obtained from 27 IGS stations as reference data. Each model is provided with 72 h of historical ZTD inputs to forecast ZTD values at lead times of 3, 6, 12, 24, 36, and 48 h. The predictive performance of the SSMB-ZTD model is evaluated against several baseline models, including RNN, LSTM, GPT-3, Transformer, and Informer. The results show that SSMB-ZTD consistently outperforms RNN, LSTM, and GPT-3 in all prediction scenarios, with average improvements in RMSE reaching 31.2%, 37.6%, and 48.9%, respectively. In addition, compared with the Transformer and Informer models based on the attention mechanism, the SSMB-ZTD model saves 47.6% and 21.2% of the training time and 38.6% and 30.0% of the prediction time on average. At the same time, the accuracy is better than the two. The experimental results demonstrate that the proposed model achieves high prediction accuracy while maintaining computational efficiency in long-term ZTD forecasting tasks. This work provides a novel and effective solution for high-precision ZTD prediction, contributing significantly to the advancement of GNSS high-precision positioning and the utilization of GNSS-based meteorological information. Full article

Tumor-initiating cells (TICs) constitute a subpopulation of cancer cells with stem-like properties contributing to tumorigenesis, progression, recurrence, and therapeutic resistance. Despite their biological importance, their molecular signatures that distinguish them from non-TICs remain incompletely characterized. This study aimed to comprehensively analyze transcriptomic differences between TICs and non-TICs, identify TIC-specific gene expression patterns, and construct a machine learning-based classifier that could accurately predict TIC status. RNA sequencing data were obtained from four human cell lines representing TIC (TS10 and TS32) and non-TIC (32A and Epi). Transcriptomic profiles were analyzed via principal component, hierarchical clustering, and differential expression analysis. Gene-Ontology and Kyoto-Encyclopedia of Genes and Genomes pathway enrichment analyses were conducted for functional interpretation. A logistic-regression model was trained on differentially expressed genes to predict TIC status. Model performance was validated using synthetic data and external projection. TICs exhibited distinct transcriptomic signatures, including enrichment of non-coding RNAs (e.g., MIR4737 and SNORD19) and selective upregulation of metabolic transporters (e.g., SLC25A1, SLC16A1, and FASN). Functional pathway analysis revealed TIC-specific activation of oxidative phosphorylation, PI3K-Akt signaling, and ribosome-related processes. The logistic-regression model achieved perfect classification (area under the curve of 1.00), and its key features indicated metabolic and translational reprogramming unique to TICs. Transcriptomic state-space embedding analysis suggested reversible transitions between TIC and non-TIC states driven by transcriptional and epigenetic regulators. This study reveals a unique transcriptomic landscape defining TICs and establishes a highly accurate machine learning-based TIC classifier. These findings enhance our understanding of TIC biology and show promising strategies for TIC-targeted diagnostics and therapeutic interventions. Full article