Search

Parameterized model predictive control (PMPC) is one of the many approaches that have been developed to alleviate the high computational requirement of model predictive control (MPC), and it has been shown to significantly reduce the computational complexity while providing comparable control performance with conventional MPC. However, PMPC methods still require a sufficiently accurate model to guarantee the control performance. To deal with model mismatches caused by the changing environment and by disturbances, this paper first proposes a novel framework that uses reinforcement learning (RL) to adapt all components of the PMPC scheme in an online way. More specifically, the novel framework integrates various strategies to adjust different components of PMPC (e.g., objective function, state-feedback control function, optimization settings, and system model), which results in a synthesis framework for RL-based adaptive PMPC. We show that existing adaptive (P)MPC approaches can also be embedded in this synthesis framework. The resulting combined RL-PMPC framework provides a solution for an efficient MPC approach that can deal with model mismatches. A case study is performed in which the framework is applied to freeway traffic control. Simulation results show that for the given case study the RL-based adaptive PMPC approach reduces computational complexity by 98% on average compared to conventional MPC while achieving better control performance than the other controllers, in the presence of model mismatches and disturbances. [ABSTRACT FROM AUTHOR]

Traditional Adaptive Cruise Control (ACC) systems often struggle to dynamically adapt to rapidly changing traffic conditions, resulting in suboptimal performance. Additionally, with fuel consumption emerging as a critical consideration alongside safety, there is a pressing need for more advanced solutions. This paper presents a novel approach to address these challenges by integrating Fuzzy ACC with Model Predictive Control, denoted as FACMPC. This integration aims to enhance both the longitudinal safety of AVs and fuel efficiency by considering real-time traffic conditions. The FACMPC system utilizes fuzzy logic inside the MPC, adaptively generates controller's weighting factors, allowing the system to adapt instantly to varying traffic environments and driving circumstances. The findings show that this adaptation improves the balance between driving safety, efficiency, and comfort. Additionally, three interruption scenarios, Alpha, Beta and Gama, are examined. In Alpha, the study evaluates the sensitivity of the FACMPC to disturbances by applying band-limited white noise to the lead vehicle velocity. In Beta, the AV experiences a loss of the lead vehicle velocity signal for a defined period, prompting safety considerations and assumptions. The Gama scenario includes a sensitivity analysis to account for variations and uncertainties in parameters by considering a range of ±5% around the nominal values for four key parameters: road slope, wind speed, wind direction, and rolling resistance. The findings indicate that the proposed controller's mean fuel consumption is 8.110, only a 3.21% increase over the nominal, compared to a 7.03% increase for the conventional ACC, demonstrating greater robustness against uncertainties. [ABSTRACT FROM AUTHOR]

Short-term prediction of high-speed rail (HSR) passenger flow provides a daily ridership estimation for the near future, which is critical to HSR planning and operational decision making. This paper proposes a new methodology that integrates ensemble empirical mode decomposition with multivariate support vector machines (EEMD-MSVM). There are four steps in this hybrid forecasting approach: (i) explore the correlation of multivariate HSR passenger flows at various stations based on archived data; (ii) decompose empirical modes of historical passenger flows for each HSR station, using EEMD to generate a number of intrinsic mode functions (IMFs) and a trend term; (iii) predict the IMF for each correlated station pair using MSVM; and (iv) reconstruct the refined IMF components to predict daily multivariate HSR passenger flows. The proposed EEMD-MSVM approach is demonstrated with multiple OD pairs along the Wuhan-Guangzhou HSR in China. Results from various origin-destination pairs, show that the EEMD-MSVM approach outperforms the existing ensemble empirical mode decomposition with grey support vector machine approach (EEMD-GSVM). With the multivariate approach, the mean absolute percentage error in demand prediction is reduced by 13.9%, 1.2%, 1.0%, 2.0%, and 2.7% and the mean absolute deviation is reduced by 78.8, 38.0, 4.4, 4.6, and 3.9, between these OD pairs respectively. Such increase in short-term demand prediction accuracy can significantly improve HSR service planning, operations, and revenue management in the real world. • Multivariate forecasting for HSR passenger flows with station correlations. • HSR network stations are classified into several groups with multivariate EEMD-EEMD. • Short-term passenger flows forecasting model for a univariate OD pair. • Mode reconstruction for multi-IMFs instead of a summation of each IMF. [ABSTRACT FROM AUTHOR]

The specifications of a certain domain are addressed by a portfolio of software products, known as Software Product Line (SPL). Feature Model (FM) supports domain engineering by modeling domain knowledge along with variability among SPL. The quality of FM is one of the significant factors for the successful SPL in order to attain high quality software products. However, the benefits of SPL can be reduced due to defects in FM. Dead Feature (DF) is one of such defects. Several approaches exist in the literature to detect defects due to DF in FMs. But only a few can handle their sources and solutions which are cumbersome and difficult to understand by humans. An ontological knowledge-based method for handling defects due to DF in FMs is described in this paper. It specifies FM in the form of ontology-based knowledge representation. The rules based on first-order logic are created and implemented using Prolog to detect defects due to DF with sources as well as suggest solutions to resolve these defects. A case study of the product line available on SPLOT repository is utilized for illustrating the proposed work. The experiments are performed with real-world FMs of varied sizes from SPLOT and FMs created with the FeatureIDE tool. The results prove the efficiency, scalability (up to model with 32,000 features) and accuracy of the presented method. Therefore, reusability of DFs free knowledge enables deriving defect free products from SPL and eventually enhances the quality of SPL. [ABSTRACT FROM AUTHOR]

Solid oxide fuel cells (SOFCs) are promising clean energy technology to produce electricity with zero emission, fuel diversity, and high electrical efficiency. The dynamic behavior of SOFC systems is affected by exogenous inputs and complex internal multi-physics phenomena. This paper introduce a hybrid neural network to build a comprehensive predictive model for SOFC performance. The proposed algorithm includes residual blocks of temporal dilated convolution that are designed to hierarchically extract a high-level hidden abstraction of the temporal dependencies from a sequence of historic output, which contains paramount information about the SOFC dynamics. The influence of the exogenous inputs is extracted through a shallow neural network. The fusion of learned features is then fed to a fully connected network for performance prediction, representing a nonlinear autoregressive exogenous scheme. To train and evaluate the model's performance, experimental data from lab-scale SOFCs under various dynamic operations was collected. Having been validated over diverse operating conditions, the prediction performance of the proposed framework is benchmarked. The results demonstrate our model can more precisely capture the SOFC dynamical behavior and is computationally more efficient than the baseline models. The parallelism of the model holds the potential for real-time control, diagnostics, and optimization of SOFC systems. • A novel two-channel parallel neural network for predicting the performance of solid oxide fuel cells. • Effectively captures dynamics induced by both exogenous inputs and internal reactions. • Validated with diverse experimental data from real tubular fuel cells. • Compared with baseline dynamic neural networks for one-step-ahead prediction tasks. [ABSTRACT FROM AUTHOR]

Although additive manufacturing processes have matured in many areas, difficulties with regard to printing accuracy persist. Possible defects are, e.g. , the generation of unwanted internal pores or a lack of fusion between layers. In general, defects result in a deviation between as-planned and as-built geometries. Assessing the influence of such deviations on effective (mechanical) properties requires either experimental testing or numerical investigation, both of which are often too complex for being used in time-critical production settings. Previous work with the Finite Cell Method (FCM) has shown that image-based computational homogenization can assist in principle in quality monitoring of produced parts by providing effective mechanical properties of as-built geometries. However, conducting FCM computations online for each and every part of a series production demands a prohibitive amount of computational resources. The paper at hand suggests a remedy to this problem by using generative adversarial networks. This manuscript presents an integrated computational workflow for outlier detection and property monitoring, which in principle consists of four steps: (i) Carry out experiments under varying process conditions and collect respective (micro)-structural images. (ii) Compute effective properties for these images using FCM. (iii) Compose a generative adversarial network (GAN) training dataset from those images that meet desired effective properties. (iv) Use the GAN discriminator to detect outliers in series production. To this end, the discriminator is used as a classifier on as-built parts to judge whether an as-built structure is acceptable or defective. The viability of the approach is demonstrated on additively manufactured lattice structures whose geometry is acquired after production via computed tomography. The methodology is not only applicable for automated property monitoring but potentially also for reliability estimates of neural network-based property predictors. • Presentation of an integrated computational workflow for quality monitoring of additively manufactured structures. • Generative adversarial network-based (micro)-structure generation for additive manufacturing • Discriminator as a reliability indicator for image-based structural property monitoring. [ABSTRACT FROM AUTHOR]

Efficiency prediction is an essential prerequisite for dynamic decision to ensure safe, stable and energy-saving operation of centrifugal pumps in various industrial applications. This paper presents a novel segmented efficiency model that integrates both process data and centrifugal pump knowledge. The consistent trend in efficiency curves, as revealed by the affinity law, is leveraged: rapid increase to the best efficiency point followed by a gradual decrease at different rotational speeds. To accentuate the methodology employed, the prediction variance of a global Gaussian Process Regression (GPR) model is utilized to partition the curve into two intervals. Local GPR models are subsequently established for each interval to predict efficiency. A posterior probability index is introduced to assess data reliability by combining GPR predictive variance with the Bayesian theorem. Data subsets with higher approximations are selected as the training set, thereby enhancing the predictive performance of the model. Validation results demonstrate the superiority of the segmented model in terms of prediction accuracy and robustness, with an average root mean square error of 0.5948 for four testing sample sets, significantly lower than three alternative models. This study provides an accurate prediction model for centrifugal pump efficiency, offering valuable insights into the integration of process data and knowledge for making dynamic operation decisions in complex applications. [ABSTRACT FROM AUTHOR]

Image super-resolution is a visual task that reconstructs low-resolution images into high-resolution ones. Currently, many researchers favor applying Transformer-based methods to image super-resolution tasks, which have yielded promising results. However, due to the need to capture long-range dependencies across the entire image, existing research on Vision Transformers (ViT) for super-resolution reconstruction incurs high computational costs, thereby increasing system overhead. Additionally, some researchers have proposed methods based on manually sparse attention mechanisms. However, these approaches, which acquire receptive fields in a manner similar to traditional convolutions, do not fully exploit the advantages of Transformers in extracting global information, resulting in suboptimal reconstruction performance. To leverage the Transformer's ability to capture long-range dependencies, this paper introduces a novel network called RASHAT. In RASHAT, we propose an Adaptive Sparse Hybrid Attention Block (ASHAB). This module introduces a Bi-level Routing Attention(BRA), incorporating both Channel Attention(CA) and Switch Window Multi-head Self-attention((S)W-MSA). These components are designed to capture long-range dependencies, global context, and local dependencies within the image. Additionally, the model employs an Overlapping Cross-Attention Block(OCAB) to enhance information interaction between neighboring pixels. During model training, we introduce a novel composite loss function that combines frequency domain loss with pixel loss, further improving model performance. Extensive experiments demonstrate that benefiting from the sparse attention provided by the Bi-Level Routing Attention (BRA), RASHAT achieves similar performance to the current state-of-the-art results (20.8M) with significantly fewer parameters (11.6M). These results hold across multiple commonly used datasets. • Designed an image super-resolution network based on hybrid attention. • Implement adaptive sparse attention with BRA to capture long-distance dependencies. • Using ASHAB to simultaneously capture global, local, and long-range dependencies. • propose and use a hybrid loss function to obtain frequency domain supervision. • Our method achieves similar performance to SOTA with fewer parameters. [ABSTRACT FROM AUTHOR]

Pan–Tilt–Zoom (PTZ) cameras, with their significant features of free rotation and zoom, are widely used in areas such as border security, ecological conservation, emergency management, and the military. PTZ cameras can achieve automatic monitoring of a selected area through multiple servo operations, known as multi-point servo control. However, due to the deficiencies in the servo control Software Development Kit (SDK), hardware wear, and interference from complex external environments, the multi-point servo control process generates significant errors. This paper proposes a precise multi-point servo control framework based on Deep Reinforcement Learning (DRL) to address this issue. The complexity of real-world environments necessitates a reward function design that fully considers various factors, for which we propose the directional gravity reward function. Due to the instability during the training process, prolonged trial-and-error interactions between the agent and the equipment can cause irreversible damage to the devices. This framework employs a phased training approach, where the agent learns sequentially from offline, off-policy, and on-policy data, reducing direct interaction with the equipment while enhancing the agent's overall performance. Additionally, real-time and accurate device status can be obtained by performing feature matching on images from adjacent time frames, which is crucial for the system's operation. Evaluation results indicate that our proposed precise multi-point servo control framework significantly outperforms other methods in 4-point servo control tasks in both virtual and real-world scenarios. Additionally, the operational process fully considers safety issues. [ABSTRACT FROM AUTHOR]

This paper presents a framework to predict the entire displacement time histories of all floors of buildings using a novel double-head neural network composed of causal Convolution Neural Network layers and Long Short-Term Memory layers. State-of-the-art deep techniques have been effectively employed to construct a surrogate model as an alternative to traditional time-consuming and expert-reliant approaches in evaluating damages and response modeling. An innovative loss function is introduced, facilitating the concurrent implementation of stacking, normalization, and multi-task learning capabilities, directing the network to simultaneously learn the time-series data and its rate of variation at every time step. This approach proves beneficial, particularly in maximum displacement and residual displacement predictions, and when dealing with displacement time series predictions in highly nonlinear regions. A comprehensive database is developed including ground motions and roof floor accelerations of steel buildings with eccentrically braced frames (EBF) and moment-resisting frames (MRF) as inputs and displacement time histories of all floors as outputs. The database involves 2,210,000 and 1,092,000 time steps, for EBF and MRF resulting from 44 far-field and 40 large magnitude-small distance ground motions, respectively, through incremental dynamic analysis. The performance of the proposed method is evaluated on scaled new ground motions in different damage states. Results confirm that the proposed stacked network predicts displacement time histories of the case study buildings with 95% accuracy for unseen datasets. In addition, the suggested loss function enhances the prediction performance by a minimum of 12.5% compared to other state-of-the-art loss functions when using the same deep learning architecture advanced in this research. Finally, a graphical user interface is developed to make the proposed approach time-efficient and expert-needless. • Implements a hybrid stacked CNN-LSTM architecture to predict displacement time history of multi-story buildings. • Uses a large database of ground motions and roof accelerations for eccentrically braced and moment-resisting frame buildings. • Introduces a novel loss function with stacking, normalizing, and multi-task learning effects. • Improves the displacement time series predictions in high non-linear regions for all floors. • Develops a graphical user interface for testing the proposed network time-efficient and user-friendly. [ABSTRACT FROM AUTHOR]

Facial expression recognition (FER) is crucial in enhancing human-computer interaction. While current FER methods, leveraging various open-source deep learning models and training techniques, have shown promising accuracy and generalizability, their efficacy often diminishes in real-world scenarios that are not extensively studied. Addressing this gap, we introduce a novel in-the-wild balanced testing facial expression dataset designed for cross-domain validation, called BTFER. We rigorously evaluated widely utilized networks and self-designed architectures, adhering to a standardized protocol. Additionally, we explored different configurations, including input resolutions, class balance management, and pre-trained strategies, to ascertain their impact on performance. Through comprehensive testing across three major FER datasets and our in-depth cross-validation, we have ranked these network architectures and formulated a series of practical guidelines for implementing deep learning-based FER solutions in real-life applications. This paper also delves into the ethical considerations, privacy concerns, and regulatory aspects relevant to the deployment of FER technologies in sectors such as marketing, education, entertainment, and healthcare, aiming to foster responsible and effective use. The BTFER dataset and the implementation code are available in Kaggle and Github , respectively. [ABSTRACT FROM AUTHOR]

Drones have become a crucial element in current wildfire and smoke detection applications. Several deep learning architectures have been developed to detect fire and smoke using either colour-based methodologies or semantic segmentation techniques with impressive results. However, the computational demands of these models reduce their usability on memory-restricted devices such as drones. To overcome this memory constraint whilst maintaining the high detection capabilities of deep learning models, this paper proposes two lightweight architectures for fire and smoke detection in forest environments. The approaches use the Deeplabv3+ architecture for image segmentation as baseline. The novelty lies in the incorporation of vision transformers and a lightweight convolutional neural network architecture that heavily reduces the model complexity, whilst maintaining state-of-the-art performance. Two datasets for fire and smoke segmentation, based on the Corsican, FLAME, SMOKE5K, and AI-For-Mankind datasets, are created to cover different real-world scenarios of wildfire to produce models with better detection capabilities. Experiments are conducted to show the benefits of the proposed approach and its relevance in current drone-based wildfire detection applications. [ABSTRACT FROM AUTHOR]

In recent decades, the Archimedean triangular norm (t-norm) and Archimedean triangular conorm (t-conorm) have been fundamental theories in the design and construction of information aggregation functions. The application of weighted Archimedean t-norm/t-conorm-based aggregation functions has been widely extended to uncertain information environments, such as fuzzy sets and linguistic term sets. However, these functions are susceptible to aggregation failure when dealing with information groups that contain extreme values, leading to unreasonable aggregation results. This paper aims to address the issue of aggregation failure by developing an extended Archimedean t-norm/t-conorm-based linear hybrid aggregation framework. Firstly, the extended weighted Archimedean t-norms and t-conorms that are suitable for processing linguistic term sets are proposed. Furthermore, an aggregation contribution function is introduced to evaluate the impact of both extreme and normal values on aggregated outcomes. This function also facilitates the identification of deficiencies within existing weighted Archimedean t-norm/t-conorm-based aggregation functions. Secondly, building upon the expanded Archimedean t-norm/t-conorm as the foundational framework, a linear weighted hybrid operator is developed by employing an extreme value identification function as guidance. The rationality of this operator is validated through the utilization of previously defined aggregation contribution function. Subsequently, to consolidate improved basic uncertain linguistic information (IBULI) pairs, the proposed hybrid operator is employed for constructing an IBULI-aggregation function. Finally, a product ranking method is developed by integrating the proposed operator and incorporating a user credibility calculation-based approach for converting ratings to IBULIs. The efficacy and rationality of the proposed approach is substantiated through a case study and comparative analysis of car ranking application. • We introduce the aggregation contribution function of operators. • We propose a hybrid operator based on extended Archimedean t-norm/t-conorm. • We define the improved basic uncertain linguistic information aggregation function. • We develop a methodology for ranking products based on large-scale user ratings. [ABSTRACT FROM AUTHOR]

The acoustic levitation technology expands the possibility of non-contact mass measurements, avoiding contact contamination and loss, particularly for tiny objects. Current acoustic levitated object mass estimation methods focus on the mechanism models between object mass and oscillating frequency. However, the same object exhibits different oscillating behaviors in the acoustic field, especially for non-spherical objects, resulting in obvious variability and non-stationarity in oscillating signals, which limits the estimation accuracy and versatility of mass. To address these challenges, this paper proposes a novel multi-modal hybrid network (MMH-Net) method for accurate mass estimation of levitated objects. This method directly correlates the object's actual mass with the oscillatory signal. Firstly, to reveal the inherent patterns of the oscillating signals, the variational mode decomposition (VMD) is introduced to decompose the variable oscillating signals into different frequency mode sub-windows. Secondly, a global module and a feature refinement module are constructed to extract the long-term dependencies of the original oscillatory signals and refine the local regularity patterns of the sub-modal signals, respectively. In addition, a weighted fusion strategy is adopted to integrate global and refinement features better. The specially designed dual spectrum self-attention module can assign reasonable weights to information from different sources, facilitating the extraction of representative features and effectively suppressing negative migration. Experimental results demonstrate that the proposed network model performs outstandingly in acoustic levitated object mass estimation. The proposed method is suitable to mass measurements of spherical and non-spherical objects, further promoting the development of acoustic levitation systems in research and application fields. [ABSTRACT FROM AUTHOR]

Since the cable-driven parallel robots (CDPRs) are subject to the model uncertainty and external disturbances with high nonlinearity, classic trajectory tracking control approaches cannot guarantee a finite-time and robust controlling performance. To solve this problem, the paper proposes a novel adaptive interval type-2 fuzzy neural network non-singular fast terminal sliding mode control (AIT2FNN-NFTSMC) approach, which provides a high precision finite-time robust closed-loop control system. Firstly, a non-singular fast terminal sliding manifold is introduced and then, the fast singular free control law is designed to reach the fast finite-time convergence of the CDPR states to the desired trajectories. Meanwhile, an interval type-2 fuzzy neural network (IT2FNN) system is constructed to approximate the unknown nonlinearities and the uncertainties, and the adaptive control law is realized. The finite-time stability of the closed-loop control system is proved using the Lyapunov stability theorem. Finally, numerical simulations are reported to demonstrate the effectiveness of the proposed controller method. [ABSTRACT FROM AUTHOR]

In recent years, Convolutional Neural Networks (CNNs) have promoted the prosperity of Change Detection (CD). However, due to the intrinsic property of convolution kernel, this method cannot effectively model the long-distance dependency. The emergence of Vision Transformer (ViT) brings a new way to solve the problem. Based on ViT architecture, a novel Multi-Granularity remote sensing image Change Detection model (MGCDT) is proposed in this paper. We cascade several Local–Global Siamese Transformer (LGST) as backbone to extract local and global semantic discriminative features. In order to solve the serious problem of false detection and missing detection of feature boundary, a plug-and-play High Frequency Enhancement Unit (HFE) is proposed to replace the inflexible U-shaped structure to optimize the detection boundary. Considering the problem of multi-scale modeling of ground objects, a Multi-Scale Fusion Attention Unit (MSFA) is proposed, which integrates the flow of multi-scale information into the calculation process of self-attention. Finally, we utilize a Deep Feature Guidance Unit (DFG) to optimize the shallow detailed feature information. Extensive experiments show that, considering multi-granularity information, MGCDT outperforms the existing change detection algorithms on four remote sensing image change detection datasets. [ABSTRACT FROM AUTHOR]

Recognizing unauthorized Unmanned Aerial Vehicles (UAVs) within designated no-fly zones throughout the day and night is of paramount importance, where the unauthorized UAVs pose a substantial threat to both civil and military aviation safety. However, recognizing UAVs day and night with dual-vision cameras is nontrivial, since red–green–blue (RGB) images suffer from a low detection rate under an insufficient light condition, such as on cloudy or stormy days, while black-and-white infrared (IR) images struggle to capture UAVs that overlap with the background at night. In this paper, we propose a new optical flow-assisted graph-pooling residual network (OF-GPRN), which significantly enhances the UAV detection rate in day and night dual visions. The proposed OF-GPRN develops a new optical fusion to remove superfluous backgrounds, which improves RGB/IR imaging clarity. Furthermore, OF-GPRN extends optical fusion by incorporating a graph residual split attention network and a feature pyramid, which refines the perception of UAVs, leading to a higher success rate in UAV detection. A comprehensive performance evaluation is conducted using a benchmark UAV catch dataset. The results indicate that the proposed OF-GPRN elevates the UAV mean average precision (mAP) detection rate to 87.8%, marking a 17.9% advancement compared to the residual graph neural network (ResGCN)-based approach. • Precise detection of UAVs in dual day and night visions via optical flow fusion. • Using Graph Convolutional Network to optimize the mAP for UAV detection. • Time-varying lighting conditions and high background similarities. • Segregation of the UAV from the background to identify. [ABSTRACT FROM AUTHOR]

Industry 4.0 is of great significance for the development of oil industry. One of the pivotal steps towards achieving oil Industry 4.0 is accurately mastering the oilfield production dynamics, especially the changes of process parameters. Among many process parameters, the accurate modeling of water cut is extremely critical and difficult. In this paper, under laboratory conditions, oil–water flows with various preset water cuts are simulated. A fluid sensor, equipped with eight concave-shaped conductive electrodes, is employed to capture multi-channel measurement data, continuously recording the oil–water flow process from multiple angles. Subsequently, a novel flow behavior-heuristic deep learning model, named FBHWC model, is developed to model the relationship between the measurement data and water cut, achieving water cut measurement. The FBHWC model is guided by the complex flow behaviors of oil–water flows and consists of two key modules. Particularly, the multi-level feature fusion module focuses on the high-order feature extraction and fusion of sensor measurement data, while the multi-scale measurement module uses fully convolutional design to achieve accurate measurement of water cut. Experimental results show that the FBHWC model has excellent performance in measuring water cut, with mean square error of 0.016%. All these open up a new venue for exploring industrial multiphase flows through combining multi-electrode sensor and deep learning. [ABSTRACT FROM AUTHOR]

This paper addresses the issue of automated intersection control for connected vehicles in the presence of Denial-of-Service (DoS) attacks. DoS attacks can disrupt the vehicular communication network, leading to increased service time and additional transmission delays, thereby raising a risk of collisions at intersections. To mitigate this problem, we propose a resilient automated intersection control system that ensures safe passage of connected vehicles through non-signalized intersections. First, we establish a communication topology that captures the information transmission between vehicles using a vehicle conflict graph. This topology serves as the basis for designing a distributed control protocol that enables conflict-free cooperation at intersections, taking into account the stochastic delays caused by DoS attacks. To synthesize a resilient controller for the automated intersection control system, we employ the polytopic overapproximation technique based on the real Jordan form. This technique allows us to model the system dynamics and uncertainties, enabling the derivation of a set of sufficient conditions for system stability. By solving a series of linear matrix inequalities, we obtain the controller gains that satisfy these stability conditions. Numerical examples are presented to demonstrate the effectiveness of the proposed control method. [ABSTRACT FROM AUTHOR]

Cross-modal retrieval has become one of the hot topics in the field of multi-modal research, which receives widespread attention. Existing approaches adopted object region (word) features to generate final image (text) global feature vector without weighting or with unsupervised attention mechanism, which lead to mistake attention or weighting of object region (word) features. In this paper, a parallel weight control method based on Policy Gradient of Relationship Refinement (PGRR) is proposed for cross-modal retrieval, which utilizes self-attention mechanism to model the relationship between any local features and all local features within the modality, thereby more accurately using discrete and continuous policy gradient to estimate the weight of this local feature in the final global feature. Furthermore, PGRR transforms the existing iterative prediction weight pattern into parallel weight control, which significantly improves the training and inference efficiency of the model. Extensive experiments on MS-COCO and Flicker30K datasets demonstrate PGRR consistently outperforms state-of-the-art methods for the image-text matching. [ABSTRACT FROM AUTHOR]

Depth map stitching technology is urgently needed in varying fields such as panoramic three-dimensional reconstruction, underwater mapping, autonomous driving, and robot collision avoidance control. Existing image stitching research mainly focuses on three-primary-color images and heavily relies on feature detection quality, while depth map stitching is rarely studied due to the sparse features and low resolution. To address the above challenges, this paper proposes a color image-guided depth map stitching framework consisting of two stages: coarse depth map stitching and depth correction. In the first stage, a color-image-guided coarse depth map stitching network is proposed to accurately estimate the homography matrix, which can further eliminate parallax artifacts in the stitched depth map. In the second stage, a transformer-based depth correction network is designed, which combines saliency detection and regional matching strategies for depth correction, in order to solve the inconsistency in depth value introduced by point-to-point matching and improve the speed of depth correction to some extent. Extensive experiments demonstrate that the proposed method effectively solves the problem of disparity artifacts and mismatched depth values of the same target, and has excellent performance in depth map stitching. [ABSTRACT FROM AUTHOR]

The intelligent diagnosis of faults in industrial assets helps avoid unexpected interruptions of critical services. Thus, machine learning systems for monitoring machinery have a fundamental role in smart factories. The goal of this paper is to promote generalization by combining binary models into multiclass ones. This approach avoids using customized models, which do not perform well if the required conditions do not match during inference. Hence, instead of proposing multiple models for each particular condition, e.g., a model for each machine type, we evaluate different multiclass models where this information becomes an additional outcome along with the operating status. In our experiments, we use the Synthetic Minority Over-sampling Technique (SMOTE) to deal with imbalanced datasets and the PyCaret library to pre-select the most promising algorithms using binary classification. After selecting the binary models for combination, we evaluate two multiclass approaches, called partially- and totally-mixed, using the Malfunctioning Industrial Machine Investigation and Inspection (MIMII) dataset of machinery audio. The number of classifiers needed is reduced up to 83% with our proposal compared with binary classification used as a baseline approach. Furthermore, in the partially-mixed multiclass proposal, it is possible to obtain the correct classification of anomalous states in at least 88% of the samples for most devices. We show that there is a cost on computational resources for model training, which is compensated by the generalization gains verified using the F1-Score metric two times higher than the binary models. We follow the same idea of building more generic models with another existing dataset, ToyADMOS, which similarly contains machine operating sounds. The experiments confirm the results obtained with the MIMII dataset. [ABSTRACT FROM AUTHOR]

Due to its efficiency and flexibility, many unsupervised feature selection models based on dictionary learning have been proposed to select prominent features for improving the accuracy of data clustering and classification in pattern recognition, data mining, and machine learning. However, these models generally focus on how to investigate data distribution but ignore the issue that data may be contaminated by outliers, making the existing models ineffective. Motivated by this, we propose Joint Cauchy Dictionary Learning and Graph Learning for Unsupervised Feature Selection in this paper. This method makes an effort to incorporate dictionary learning, dual representation learning, and adaptive graph learning into a unified framework, so that both the global and local correlation information can be utilized to facilitate feature selection. Specifically, we first propose Cauchy dictionary learning to get high-quality latent representations that are resistant to data outliers. Following the high-quality representations, we re-encode latent features (features of latent representations) into the feature space and get dual representations to quantify the significance of data features in retaining the global correlation information of data. Then, the local similarity structure is adaptively learned in the projection space to quantify the significance of data features in retaining the local correlation information of data. Finally, we design an efficient optimization algorithm to optimize this model and theoretically analyze its convergence and computational complexity. The evaluation results on eight real-world datasets show that our method outperforms several state-of-the-art feature selection methods. [ABSTRACT FROM AUTHOR]

Grinding mills exhibit steady-state behaviour, formalised by quadratic performance functions called grind curves. Grind curves are predominantly affected by feed-ore characteristics that are generally unknown during regular operation. This work assumes that optimising grinding mills, using grind curves, is akin to bandit problems that can be solved using continuum-armed bandit algorithms. This paper proposes a constrained continuum-armed bandit agent based on Thompson sampling that uses linear models to model the agent reward as a combination of real-value and binary components. Real-value reward components represent mill-performance indicators and are modelled as second-order polynomials of the agent's actions, thereby approximating the unknown grind curves. Binary components indicate constraint violations in the environment and are used to train linear constraint models for the optimisation algorithm selecting the optimal actions. Agent testing occurred in an environment containing a dynamic grinding mill model calibrated to grind curve data. Optimisation performance was compared to extremum-seeking controllers operating in the same environment. The experiments demonstrate that the continuum-armed bandit agent could find the unknown grind curves and steer the process near the optimal operating point. The continuum-armed bandit agent followed similar trajectories to the extremum-seeking controllers in most experiments. The agent requires tuning a smaller number of hyper-parameters than the extremum-seeking controllers, with physically meaningful interpretations. The agent improved upon the optimisation performance of one variant of the extremum-seeking controllers by 2% while producing a 1% decrease compared to the remaining two controllers. The agent proved robust to algorithm hyper-parameters, obtaining similar rewards for various agent configurations. [ABSTRACT FROM AUTHOR]

In the field of cross-modal retrieval, single encoder models tend to perform better than dual encoder models, but they suffer from high latency and low throughput. In this paper, we propose a dual encoder model called BagFormer that utilizes bag-wise late interaction mechanism to improve re-rank performance without sacrificing latency and throughput. BagFormer achieves this by employing a bagging layer, which facilitates the transformation of text to an appropriate granularity. This not only mitigates the issue of modal granularity mismatch but also enables the integration of entity knowledge into the model. Our experiments have shown that BagFormer ViT-B outperforms the traditional dual-encoder model CLIP ViT-B by 7.97% in zero-shot settings. Under fine-tuned conditions, BagFormer ViT-B demonstrates an even more significant improvement of 17.98% over CLIP ViT-B. Moreover, BagFormer not only matches the performance of cutting-edge single-encoder models in cross-modal retrieval tasks but also provides efficient inference processes characterized by lower latency and higher throughput. Compared to single-encoder models, BagFormer can achieve a speedup ratio of 38.14 when re-ranking individual candidates. Code and models are available at github.com/howard-hou/BagFormer. [Display omitted] • BagFormer introduces bag-wise interaction for cross-modal retrieval. • Reduces modal granularity mismatch, improving dual encoder performance. • Achieves state-of-the-art results with lower latency and higher throughput. • Utilizes bagging layer to transform text for better image–text alignment. • Outperforms other dual encoder models in retrieval metrics. [ABSTRACT FROM AUTHOR]

With the rapid development of intelligent vehicles, mobile robots, warehousing and logistics, exploration and testing and other industries, multi-platform collaborative work research has become the future research focus of ground platforms. This paper mainly studies the multi-ground platform collaborative system as follows: Combined with the multi-vehicle collaborative scenario, a three-layer system Artificial Intelligence control framework is proposed, which is composed of a task assignment layer, trajectory planning layer and multi-vehicle control layer. Firstly, the multi-car task assignment algorithm based on the Dubins curve is studied. Secondly, the vehicle motion planning algorithm based on the D*Lite algorithm is designed, and the driving direction and the optional grid range are improved. Then, the obstacle scene of multi-vehicle formation is analyzed, two obstacle avoidance schemes of formation are proposed and the process analysis is carried out. Finally, the application in engineering is conducted to verify the effectiveness of obstacle avoidance strategy, trajectory planning strategy and the overall system framework. • A task solving algorithm based on Dubins curve is designed. • An improved D* algorithm is adopted to realize path planning. • A formation algorithm based on model prediction is designed. • Unmanned vehicles are designed and verified by experiment. [ABSTRACT FROM AUTHOR]

Beijing has long suffered from the events of fine particulate matter pollution. Consequently, it is of paramount importance to be able to scientifically and reasonably predict the concentration of particulate matter. To this end, a novel prediction model with an adaptive grey generation operator and fractional derivative is proposed. The novel model initially expands the range values for the order of the grey generation operator to enhance data preprocessing capabilities. Meanwhile, the introduction of fractional derivatives enhances the model's adaptability and overcomes the limitations of traditional grey prediction models that use integer-order differential equations. Subsequently, the fractional derivative and operator orders of the novel model are optimized to find the best parameter combination, thereby enhancing the model's performance. Furthermore, the novel model is compatible with seven other models when its parameters take different values, thus expanding the model's applicability. The novel model is then applied to simulate and predict the particulate matter concentration in Beijing, achieving a comprehensive mean relative percentage error of only 1.86%, significantly outperforming several similar models(17.56%,17.64%,3.42%). Finally, the paper presents a forecast of the particulate matter concentration in Beijing over the next four years. [Display omitted] [ABSTRACT FROM AUTHOR]

In the intelligent fault diagnosis of rolling bearings, transfer learning methods extend the applicability of models to diverse working scenarios. However, real-world scenarios often suffer from data imbalance, which reduces diagnostic accuracy. To address this issue, this paper proposes a multi-domain adversarial transfer (MDAT) framework to enhance cross-domain fault diagnosis accuracy for rolling bearings under imbalanced data conditions. First, an enhanced information generation method is introduced to produce realistic and useable synthetic data to mitigate data imbalance. Subsequently, an adversarial multi-domain adaptation module is designed to learn invariant features across multiple domains. Finally, a domain reweighting method is proposed to improve domain alignment and enhance domain confusion. The effectiveness of the proposed method was validated through two case studies on rolling bearing fault diagnosis. The results demonstrated that MDAT achieved cross-domain diagnosis accuracies of 89.2% and 99.0% under imbalanced data conditions, confirming the effectiveness and superiority of the MDAT framework. [ABSTRACT FROM AUTHOR]

The effectiveness of knee rehabilitation systems in aiding patients with rehabilitation training has been well-documented. Presently, there is an increasing emphasis on the wearing comfort and user engagement of these systems. In this paper, a wearable knee rehabilitation system (WKRS) based on the graphene textile composite sensor (GTCS), which subjects perform rehabilitation training by controlling the ascent and descent of a bird in the visual feedback game utilized GTCS, is proposed and investigated. To obtain an accurate and smooth estimated knee joint angle, we propose an improved tree boosting regression algorithm based on extreme gradient boosting (XGBoost), called improved XGBoost (IXGB). Specifically, the estimated results are smoothed by two steps: curving the preliminary estimating results within the same interval and smoothing the preliminary estimating results using the weighted moving average method. An online experiment with ten subjects validates the effectiveness of the WKRS and IXGB. Results indicate that the IXGB significantly enhances the smoothness of estimated value while maintaining estimation accuracy compared to XGBoost, random forest regression (RFR) and support vector regression (SVR). IXGB achieves an average increase of 33.43 % in root mean square error, 23.83 % in R-squared, and a 62.63% decrease in smoothness. [ABSTRACT FROM AUTHOR]

Accurately forecasting financial time series in the dynamic shipping market is vital for stakeholders, such as shipowners, investors, brokers, and shipyards. This paper introduces an innovative hybrid machine learning model leveraging Light Gradient Boosting Machine (LightGBM) to enhance financial time series predictions within the international shipping sector. LightGBM, known for its efficiency and scalability in handling high-dimensional data, offers a robust foundation for this forecasting endeavor. However, LightGBM fails to extract temporal features from time series data. Time series usually contain multi-scale information with different frequencies, but LightGBM directly learns from the original time series, and the forecasting accuracy is unsatisfactory. To address this challenge, we propose a two-stage hybrid forecasting model. In the initial stage, we employ the variational mode decomposition method to extract predictive features from the time series data online. Subsequently, we employ LightGBM for forecasting purposes, capitalizing on its superior capabilities. To validate the effectiveness of our approach, we conduct an extensive empirical study involving sixteen distinct time series from the global shipping market. We illustrate the model's superiority in forecasting accuracy and reliability through comprehensive comparisons with state-of-the-art methods from the existing literature. This research provides valuable insights for stakeholders and showcases the potential of hybrid machine learning techniques for financial time series forecasting in complex and dynamic markets. [Display omitted] • A hybrid machine learning model leveraging LGBM was proposed to enhance time series forecasting. • Forecasting accuracy and reliability were presented to show the model's superiority. • The variational mode decomposition method was applied to select input features. [ABSTRACT FROM AUTHOR]

Clustering has a wide range of applications in various industries. In different research ideas, bionic algorithms have heuristic help for clustering research. However, most of the clustering algorithms based on bionics often have problems of insufficient information extraction, too many required parameters, and limited application scope. Therefore, these algorithms have lower accuracy than traditional clustering algorithms. To solve these problems, this paper proposes a new ant colony clustering algorithm, which is called Dyeing Clustering Algorithm based on Ant Colony Path-finding. This algorithm combines the idea of dyeing with pheromones for the first time and realizes clustering by using ants to crawl and dye each point, bringing new inspiration to the artificial intelligence research. This can better capture data features, improve fault tolerance and achieve a better clustering effect. In addition, the algorithm is constructed as a dynamic system to deeply mine data information with the strategy of swarm intelligence, including the ant colony generation model, path-finding model, path-finding termination model, and dyeing clustering model. Besides, to evaluate the algorithm, we selected 2 simulation data sets, 5 artificial data sets, 5 real-world data sets, 13 comparison algorithms, and 8 evaluation indexes. From the comprehensive multi-round experiments, seven of the eight evaluation indexes of the Dyeing Clustering Algorithm based on Ant Colony Path-finding algorithm are more advantageous than other comparison algorithms, which demonstrates its enormous potential in industrial applications. [ABSTRACT FROM AUTHOR]

Reconstructing gene regulatory networks (GRNs) is an important task to reveal the regulatory relationship between genes and understand the mechanism of intracellular gene expression regulation. With the development of single-cell ribonucleic acid sequencing (scRNA-seq) technology, researchers begin to attempt to infer GRN within cells. In this paper, we propose graph attention network with convolutional layer (GATCL) to infer the latent interactions between transcription factors (TFs) and target genes in GRN. Firstly, GATCL uses graph attention network (GAT), which can effectively extract information about genes and TFs. Secondly, we combine multi-head attention layer with one-head attention layer and propose a new method of using convolution instead of weight matrix. Thirdly, we use the exponential linear unit (ELU) activation function to replace the leaky rectified linear unit (LReLU) commonly used in GAT, which further improves the accuracy of GATCL. The AUROC of our method on seven scRNA-seq datasets with four types of ground-truth networks reached an average of 0.827, which is higher than other state-of-the-art models. GATCL applies a supervised deep learning algorithm to solve the problems existing in the inference of GRN from scRNA-seq in the engineering field, and the effectiveness of this method is verified by a large number of experiments. [ABSTRACT FROM AUTHOR]

Wildfires are complex phenomena with harmful consequences, ranging from environmental and property destruction to loss of human lives. In this sense, predicting wildfire behaviour is essential to mitigate its impacts and consequences. The Rothermel model is the most used fire rate of spread prediction model. However, input parameter uncertainty is a significant source of prediction error. In this paper, we propose the calibration of the input parameters of the fire propagation model by metaheuristic algorithms under a two-stage framework. The fire spread model consists on the Rothermel model in a two-dimensional approach for surface fires. The proposed calibration is performed in two stages iteratively repeated over time: (i) the calibration of the fire spread model's input parameters and (ii) the wildfire spread prediction using the calibrated input parameters. The calibration was performed by the genetic algorithm, differential evolution, and simulated annealing, which calibrates the surface-area-to-volume ratio, fuel bed depth, live fuel moisture and dead fuel moisture. The symmetric difference between the real and predicted fire map shapes was defined as the fitness function of all three metaheuristic algorithms. For validation, simulations were done on two prescribed fires. The results for the real and estimated fire behaviour were then compared and revealed that all the tested metaheuristic algorithms produce a better fit to the real fire's perimeter when compared to the uncalibrated Rothermel model. From the results, differential evolution provided the majority of best results when compared to genetic algorithm and simulated annealing algorithms in each scenario. [ABSTRACT FROM AUTHOR]

This paper firstly deals with axiomatic definitions of some significant information measures of PFSs (Pythagorean fuzzy sets), including inclusion measure, similarity measure, knowledge measure, distance measure and entropy measure. Afterwards we investigate the transformation relationship among them for PFSs. Simultaneously, some new specific formula to compute information measure of PFSs are put forward. In addition, appropriate examples are provided for each developed transformation to support these processes. Later, three comparisons with some existing Pythagorean fuzzy information measure are constructed to illustrate their superiority. Finally, we respectively apply information measure in medical diagnosis and MCDM (multi-criteria decision-making) to state their effectiveness. [ABSTRACT FROM AUTHOR]

Unmanned aerial vehicles (UAVs), particularly quadrotors, have seen a surge in applications ranging from surveillance to delivery services. However, autonomous control remains a challenge due to the complexity of their dynamics and the need for real-time responsiveness. This paper addresses the significant problem of autonomous quadrotor control by introducing a novel learning method that integrates a deep neural network with advanced control techniques. Novelty of the work is distribution of different learning tasks in different layers of learning and combining the total learning system in a applicable UAV. The purpose is to enhance the quadrotor's ability to perform complex maneuvers autonomously, which is crucial for expanding its application scope. The proposed method begins with the acquisition of direction control using a fractional order proportional–integral–derivative (FOPID) controller. The learning parameters are refined through a fuzzy method, ensuring adaptability and precision. For the optimization of the FOPID parameters, we apply two stochastic algorithms: Genetic Algorithm and Particle Swarm Optimization, comparing their efficacy. Our approach successfully teaches the quadrotor to generate and control a flight path autonomously. The second stage of learning involves proximal policy optimization, enabling the execution of intricate maneuvers. The experimental setup includes a quadrotor on a test stand with three degrees of freedom, where we validate the method through rigorous simulation and physical tests. The results demonstrate that our method not only teaches the quadrotor complex maneuvers but also ensures robust and precise execution, marking a significant advancement in UAV control systems. The implications of this research extend to improving UAV reliability for critical applications, potentially transforming the landscape of autonomous aerial operations. [ABSTRACT FROM AUTHOR]

This paper proposes a new, novel ranking method called Inverse-Reinforcement Learning Ranking. The main goal is to find a reward function representing the user's perceived utility after clicking on each result. It is necessary to utilize log information of all users' queries in the search engine dataset to reach this goal while assuming that the decisions (clicks) of the users are the best (optimal policy). The respective reward function is constructed using features extracted through a feature selection, and their corresponding weights are obtained by two optimization models, which are applicable for ranking results represented to the new users. In addition, new performance criteria were developed to illustrate the performance of the presented ranking method. To evaluate and test the proposed ranking algorithm, a real medium-sized dataset from a search engine is preprocessed and used in this research. Findings show promising results and decisive superiority over the default ranking method. It is illustrated that clicks on the top five results, top ten, and even the first results are remarkably improved by about 13–19% in all experiments, and the perplexity remarkably decreases by almost 23% after applying the ranking method. [ABSTRACT FROM AUTHOR]

Reinforcement learning (RL) has emerged as a powerful tool for addressing complex decision making problems in various domains, including aviation. This paper provides a comprehensive overview of RL and its applications in the aviation industry. We begin by introducing the fundamental concepts and algorithms of RL, highlighting their unique advantages in learning from interaction and optimizing decision-making processes. We then delve into a detailed examination of the successful implementation of RL methods in aviation, covering areas such as flight control, air traffic management, airline revenue management, aircraft maintenance scheduling, etc. Furthermore, we discuss the potential benefits of RL in enhancing safety, and sustainability within the aviation sector. Finally, we identify and explore open challenges and areas for future research, emphasizing the need for continued innovation and collaboration between the fields of reinforcement learning and aviation. [ABSTRACT FROM AUTHOR]

Defect detection holds significant importance in industrial manufacturing as it aims to improve production quality and efficiency by identifying surface defects in products. Although previous works have shown excellent performance on many common benchmark datasets, they are often limited in terms of generality and their ability to detect logical anomalies. In this paper, we propose a dual-branch reverse distillation framework specifically designed for screw detection. By extracting global information through the global branch and combining it with local information, this framework enables the simultaneous detection of both logical and structural anomalies. To recover the information lost in compressed feature representation and improve the reconstruction ability of the decoder, we introduce a novel decoder structure that integrates multiple additional image input modules into the original network, allowing for various combinations of composite information with the input image during the image reconstruction process. Experimental results demonstrate the effectiveness of our proposed method on benchmark datasets. [ABSTRACT FROM AUTHOR]

It is challenging to achieve compliant operation in tight interaction tasks, where closed-loop constraints are formed between the robots or between a robot and the environment. Complex dynamic interactions and uncertain parameters degrade the performance of model-based controllers. In this paper, a reinforcement learning-based distributed impedance control approach is proposed for these tight interaction tasks. Two aspects are considered to ensure compliant operation for the robots. First, a distributed impedance model is established through the design of reasonable independence states and nodes networks. Second, the reinforcement learning agent is designed to make decision for adjusting impedance parameters. The trained parameters are integrated into the designed impedance model, and a model-based controller is then employed for compliant control for the robots. The effectiveness is validated under two different tight interaction scenarios via co-simulation. Compliant operation can be achieved whether between robots or between a robot and the environment. • An RL-based distributed impedance method is proposed for force-position compliance control of robot in tightly interacting tasks, where a closed-chain contact should be maintained between the robots or between the robot and the environment. • The designed distributed impedance model can describe the dynamic characteristics. Whether it is a single robot or a multi-robot closed-loop constraint system, this makes the method more extensible. Additionally, the integration of the RL agents enable better fitting of impedance parameters. • From a practical point of view, different scenarios and various factors are considered in simulations. The robot can accomplish the desired tightly interacting tasks in a co-simulation platform between MATLAB and Coppeliasim. The results demonstrate the effectiveness of the proposed RL-based distributed impedance control method. [ABSTRACT FROM AUTHOR]

Fundus examination, conducted through the analysis of retinal images, plays a pivotal role in aiding the diagnosis of ophthalmic diseases. Currently, deep learning models have been widely applied to retinal image analysis, specifically retinal image segmentation, including vessels, optic cups, lesions, etc., yielding promising outcomes. Nonetheless, the performance of these deep learning models experiences substantial degradation due to the domain shift between the distribution of the training images and unseen test images. In this paper, we focus on the challenging single-domain generalization (SDG), which aims to learn a generalized model on only one source domain, with the expectation that it performs well on unseen test domains. Our work is motivated by the observation that the main differences in retinal images from different domains primarily reside in their color variations, rather than changes in the shapes of objects. To this end, we present random color transformation (RCT) for SDG. RCT performs random linear transformations to each color channel of the training image. Through this approach, RCT can generate training images with rich and vibrant color representations, while preserving the structural information of objects in the images. Experiments are conducted over optic cup segmentation, retinal vessel segmentation, and diabetic retinopathy multi-lesion segmentation tasks, involving eight publicly available datasets. Experimental results show that the proposed RCT outperforms comparison SDG methods, achieving improvements of 5.9%, 1.3% and 3.6% compared to the second-best method on the optic cup, vessel, and lesion segmentation tasks, respectively. The source code will be available at https://github.com/guomugong/RCT. [ABSTRACT FROM AUTHOR]

To increase the economic efficiency of utilizing wind turbines, an enhanced object detection model based on the small version of the fifth generation of the You Look Only Once algorithm (YOLOv5s) is proposed in this paper, which can effectively detect cracks on the surface of wind turbine blades using images captured by unmanned aerial vehicles. To improve the extraction of light-colored and low-definition images, Omni-Dimensional Dynamic Convolution (ODConv) and Dynamic Head Module components (DyHead) are introduced in the proposed method, while a lightweight Group Shuffle convolution (GSConv) module is utilized to accelerate the model inference speed without sacrificing detection performance. Furthermore, a semi-supervised learning strategy is developed to reduce human labors in annotating images. Extensive experiments demonstrate that the proposed model outperforms the original YOLOv5s in terms of both detection accuracy and inference speed. Besides, the proposed model has good performance against state-of-the-art methods. Furthermore, the experiments validate the efficacy of the proposed semi-supervised learning strategy. [ABSTRACT FROM AUTHOR]

The inherent nonlinear and memory-dependent input-output characteristics of piezoelectric actuators pose challenges to the precision of piezoelectric positioning systems. In order to solve this problem, this paper firstly transforms the Jiles-Atherton (JA) model into a neural network structure, designs the Jiles-Atherton neural network (JANN), and combines JANN with nonlinear autoregressive exogenous input (NARX) neural network. A hybrid JA-NARX neural network model is proposed for the first time. This model has the advantages of simple structure, high modeling accuracy, and good interpretability. The effectiveness of the proposed JA-NARX neural network model is validated through a series of experiments, specifically assessing its capacity to accurately capture rate-dependent and asymmetric hysteresis characteristics. The results show that although the proposed neural network model has fewer layers and relatively simple structure, it can realize the high-precision modeling of piezoelectric hysteresis dynamics at a lower computational cost. The experimental data shows that, under the excitation of 60 Hz input signal, the model's PV error only accounts for 0.82% of the full scale range, and the modeling performance is far superior to other models. [Display omitted] • A shallow neural network based on Jiles-Atherton differential-equation model is proposed and constructed for the first time. • An innovative JA-NARX hybrid neural network model is proposed by combining Jiles-Atherton with NARX models. • The JA-NARX hybrid neural network model features simple structure, high modelling accuracy, strong interpretability. • The proposed JA-NARX model was rigorously validated through experiments, effectively describing complex dynamic hysteresis. [ABSTRACT FROM AUTHOR]

The aviation safety reporting system (ASRS) data, which are accident investigation reports issued by the national transportation safety board (NTSB), are often used for aviation risk level identification modeling. However, ASRS data suffer from sample imbalance and inaccurate encoding of textual information. To address these issues, this paper proposes a novel risk identification model. Initially, the pre-trained sentence bidirectional encoder representations from transformers (SBERT) model is fine-tuned using the textual information in the ASRS data, improving the accuracy of the text encoding. Subsequently, a generative adversarial network (GAN) is employed to generate samples for underrepresented categories in the ASRS data, thereby addressing the sample imbalance problem. Finally, a bayesian optimization algorithm is integrated to automatically search for the hyperparameters of the risk identification model, further enhancing the model's performance in risk level identification. Experimental results demonstrate that not only does the SBERT fine-tuning substantially improve the accuracy of text information encoding, but both data augmentation and automatic hyperparameter search also significantly contribute to the performance improvement of the risk level recognition model. [ABSTRACT FROM AUTHOR]

This paper proposes a method to compute the solutions of linear optimal control expressed in terms of the radial basis function neural networks with Gaussian activation functions for multi-degree-of-freedom dynamic systems. Hamilton–Jacobi–Bellman equation is adopted to formulate the optimal control problem. The radial basis function neural networks are proposed to approximate the value function to solve the Hamilton–Jacobi–Bellman equation with a policy iteration algorithm. A dominant stabilizing control is proposed as an initial control to start the policy iteration that guarantees the convergence of the iteration, particularly for open-loop unstable dynamic systems. The balanced truncation technique is applied to the multi-degree-of-freedom dynamic system to reduce the dimension of the original system, which provides multiple advantages when applying the radial basis function neural networks to unstable dynamic systems in a relatively high dimensional state space. Transfer learning is also adopted to update radial basis function neural networks with experimental data, which results in further control performance improvement. Numerical simulations and experimental studies show that the radial basis function neural networks not only find accurate optimal controls for linear systems, but also offer excellent performance in trajectory tracking and stabilization applications. [ABSTRACT FROM AUTHOR]

The detection of series arc faults is crucial due to the potential safety hazards caused by cable aging and breakage. This paper proposes an arc fault detection framework that utilizes the relative position matrix and deep convolutional network to accurately detect alternating current series arc faults. In this framework, the time series data is first converted into a relative position matrix and then visualized as a two-dimension image. This enables the characterization of singular values and flat shoulders in the arc current, making it easier to extract significant features for the models. Next, a Residual Network with hybrid attention mechanisms model is constructed and the images are fed into the model to determine whether an arc fault has occurred in the current signal. Finally, the data is collected using the experimental platform, and the relative position matrix is constructed with different dimensionality reduction factor. The appropriate dimensionality reduction factor is selected through experimental comparisons. In addition, the effectiveness of the proposed method is verified by measured data, and the proposed model is compared with four other commonly used network models, which demonstrates the superiority of the proposed model in terms of detection performance. The proposed method achieves a detection accuracy of 98.74%. • Time series data is converted to relative position matrix to enhanced fault features. • Integrating the hybrid attention mechanism to extract the critical feature. • A new network is proposed to achieve high-precision and fast arc fault detection. • The effectiveness of proposed framework has been fully validated. [ABSTRACT FROM AUTHOR]

Electroencephalography (EEG) motor imagery (MI) signals have recently attracted much attention because of their potential to communicate with the surrounding environment in a specific way without the need for muscular and physical movement. Despite these advantages, these signals are difficult to detect due to their low signal to noise (SNR) rate and non-stationary and dynamic nature. Convolutional neural networks (CNN) can extract appropriate spatial and temporal features without the need for separate feature extraction and classification steps. Clean input data for CNN significantly improves its performance, but sometimes, common interference occurs between multi-channel signals. This challenge, along with the noisy EEG signals, degrades the performance of CNN networks. This paper presents a channel selection method based on convolutional neural networks to obviate these challenges. The proposed shallow convolutional neural network (SCNN) is designed with temporal convolution and pointwise convolution (using 1 × 1 convolution) to select the best channel with minimal computational load. Following the channel selection step, the multi-layer fusion CNN model was used to classify the signals. Choosing suitable and relevant EEG channels for motor imagery tasks makes it possible to create appropriate inputs for the classification model so that the model can be improved in the end. For High Gamma Dataset and Brian-Computer Interface (BCI) Competition IV-2a, the following accuracies were obtained: 81.15% and 72.01%, which are higher than when the other channel selections such as Mutual Information, Sequential Feature Forward Selection (SFFS) and wrapper methods are used. [ABSTRACT FROM AUTHOR]

Due to the advantages of indicating variable structure and efficient reasoning, Bayesian Networks (BN) have been widely used in data-driven soft sensor applications. However, restricted to linear and conditional Gaussian property, BN-based soft sensors rarely achieve high prediction accuracy. In this paper, an ensemble learning framework – additive dynamic Bayesian networks (ADBN) is proposed for enhanced feature learning, in which Dynamic Bayesian networks are used to learn the conditional independencies among variables and construct feature sets for the following base learners. Additional DBNs are constructed upon the residual information from the past model, to carry out feature learning to fit the residuals. The procedure is repeated and a termination rule from the perspective of feature learning is proposed to end this process, and thus the model complexity can be well restricted. The proposed method is validated on two actual industrial cases. It reveals that the ADBN feature learning method has obtained great improvements. Compared to the single DBN feature engineering method, the root mean square error (RMSE) performance has been improved by 20% and 13% on the two cases, respectively. [ABSTRACT FROM AUTHOR]

Pointer meters are widely used in various industries, and there is a growing demand for automatic and non-intrusive access to meter readings. The existing methods of automatically calculating meter readings involve complex processes which are time-consuming and have low automation performance. The paper presents the design of a Separate Dual-Selective Attention Mechanism (SDSAM) module and a Reading Correction Module (RCM). We also introduce the SDSM-DenseNet model, which utilizes dense connections to directly predict meter readings based on meter images. The SDSM-DenseNet model demonstrates high automation performance and reduces reliance on inherent characteristics of the meter. Compared to other attention mechanism models and feature extraction models, the SDSM-DenseNet model achieves lower error rates in calculating meter readings. Furthermore, when compared with representative methods for reading meters, the average error rate of the SDSM-DenseNet model is reduced by approximately 48%, while only requiring 0.021 s to predict a single image. [ABSTRACT FROM AUTHOR]

In recent years, there has been a growing interest in analyzing text data from different scientific fields. The significant advancement of Artificial Intelligence in Natural Language Processing enables a systematic categorization of the wealth of scientific papers into fundamental thematic clusters. In this context, topic modeling is playing a crucial role. Unfortunately, the comparative analysis between traditional and advanced topic modeling methods, including well-established techniques like Latent Dirichlet Allocation (LDA) and newer approaches like BERTopic, remains significantly underexplored. This study addresses this gap by conducting a comprehensive analysis of extensive text data focused on sustainable energy research. To achieve this, we compile a unique dataset consisting of thousands of abstracts sourced from PubMed, Scopus, and Web of Science. Our analysis involves a comparison between LDA and the transformer model BERTopic. Importantly, we introduce a novel approach to determine the optimal number of topics, achieved through the maximization of combined semantic scores, and show that the number of topics is considerably lower than from previous approaches. Overall, our study not only contributes methodologically but also enhances our understanding of the principal topics in sustainable energy research. [ABSTRACT FROM AUTHOR]

The reliable operation of autonomous vehicles heavily depends on the functionality of numerous sensors that facilitate environmental perception and vehicle control. However, sensor malfunctions can significantly undermine the dependability and safety of autonomous vehicles. Despite the significant advancements made in sensor fault diagnosis through deep learning, the technology's susceptibility to noise and limited frequency analysis capacity renders it less effective in addressing frequency aliasing and noise interference issues. In addition, repairing incorrect sensor data holds greater significance for autonomous vehicles than merely identifying sensor failures. To this end, this paper proposes a wavelet-powered hierarchical frequency filtering framework (WavePHF) for autonomous vehicle sensors' fault diagnosis and correction. First, this study designs a multi-sensor fault identification and correction joint architecture to detect sensor faults and repair fault signals in time. Second, this study integrates the discrete wavelet transform (DWT) algorithm into the deep model to significantly enhance the model's frequency analysis capabilities and suppress frequency aliasing and noise interference. In particular, DWT is deeply integrated into the convolutional modules' parameter updating and optimization process, thereby facilitating mutual learning and promotion. Third, this study proposes an adaptive frequency weighting block for low-frequency features and an adaptive frequency discarding block for high-frequency features to achieve hierarchical frequency filtering in a coarse-to-fine manner and efficiently identify fault categories and restore pure signals. The WavePHF is evaluated on a real autonomous driving dataset. Experiments show that WavePHF has excellent autonomous vehicle sensors fault diagnosis and correction capabilities and competitive performance in different urban scenarios. [ABSTRACT FROM AUTHOR]