Search Results (41,666)

Light detection and ranging (LIDAR) sensors using a polarization-diverse receiver are able to capture polarimetric information about the target under measurement. We demonstrate this capability using a silicon photonic receiver architecture that enables this on a shot-by-shot basis, enabling polarization analysis nearly instantaneously in the point cloud, and then use this data to train a material classification neural network. Using this classifier, we show an accuracy of 85.4% for classifying plastic, wood, concrete, and coated aluminum. Full article

The design of the high-speed on/off valve is challenging due to the interrelated structural parameters of its driving actuator. Hence, this study proposes a multi-objective optimization approach that integrates a backpropagation neural network and artificial fish swarm algorithm optimization techniques to accurately model the electromagnetic solenoid structure. The backpropagation neural network is fitted and trained using simulation data to obtain a reduced-order model of the system, enabling the precise prediction of the system’s output based on the input structural parameters. By employing the artificial fish swarm algorithms, with optimization objectives focusing on the valve’s opening and closing times, a Pareto optimal solution set comprising 30 solutions is generated. Utilizing the optimized structural parameters, a prototype is manufactured and an experimental setup is constructed to verify the dynamic characteristics and flow pressure drop. The high-speed on/off valve achieves an approximate opening and closing time of 3 ms. Notably, the system output predicted using the backpropagation neural network (BPNN) exhibits consistency with the experimental findings, providing a reliable alternative to mathematical modeling. Full article

An accurate estimation of the state of health (SOH) of Li-ion batteries is critical for the efficient and safe operation of battery-powered systems. Traditional methods for SOH estimation, such as Coulomb counting, often struggle with sensitivity to measurement noise and time-consuming tests. This study addresses this issue by combining incremental capacity (IC) analysis and a novel neural network, Kolmogorov–Arnold Networks (KANs). Fifteen features were extracted from IC curves and a 2RC equivalent circuit model was used to identify the internal resistance of batteries. Recursive least squares were used to identify the parameters of the equivalent circuit model. IC features and internal resistance were considered as input variables to establish the SOH estimation model. Three commonly used machine learning methods (BP, LSTM, TCN) and two hybrid algorithms (LSTM-KAN and TCN-KAN) were used to establish the SOH estimation model. The performance of the five models was compared and analyzed. The results demonstrated that the hybrid models integrated with the KAN performed better than the conventional models, and the LSTM-KAN model had higher estimation accuracy than that of the other models. The model achieved a mean absolute error of less than 0.412% in SOH prediction in the test and validation dataset. The proposed model does not require complete charge and discharge data, which provides a promising tool for the accurate monitoring and fast detection of battery SOH. Full article

In response to energy consumption and traffic efficiency reduction caused by intersection congestion, a global speed planning that considered both ecological speed and green wave speed was conducted under the vehicle–road–cloud integration framework. After establishing an instantaneous energy consumption model for pure electric vehicles, a radial basis neural network model was used to estimate the queue length of traffic flow, and an isolated-intersection-based eco-approach and departure (I-EAD) plan was proposed based on a valid traffic signal light model. A two-stage optimization multi-intersections-based eco-approach and departure (M-EAD) strategy with multiple objectives and constraints was proposed to solve the optimal green light window and the optimal speed trajectory. The results of the SUMO/Matlab/Simulink/Python joint simulation platform show that the M-EAD strategy reduces the average travel energy consumption by 16.65% and 8.31%, and the average travel time by 26.33% and 12.53%, respectively, compared to the intelligent driver model (IDM) and I-EAD strategy. The simulation results of the typical traffic scenarios and random traffic scenarios indicate that the speed optimization strategies in this study have good optimization effects on energy conservation and traffic efficiency. Full article

Parkinson’s disease (PD), the second most prevalent neurodegenerative disorder worldwide, frequently presents with early-stage speech impairments. Recent advancements in Artificial Intelligence (AI), particularly deep learning (DL), have significantly enhanced PD diagnosis through the analysis of speech data. Nevertheless, the progress of research is restricted by the limited availability of publicly accessible speech-based PD datasets, primarily due to privacy concerns. The goal of this systematic review is to explore the current landscape of speech-based DL approaches for PD classification, based on 33 scientific works published between January 2020 and March 2024. We discuss their available resources, capabilities, and potential limitations, and issues related to bias, explainability, and privacy. Furthermore, this review provides an overview of publicly accessible speech-based datasets and open-source material for PD. The DL approaches identified are categorized into end-to-end (E2E) learning, transfer learning (TL), and deep acoustic feature extraction (DAFE). Among E2E approaches, Convolutional Neural Networks (CNNs) are prevalent, though Transformers are increasingly popular. E2E approaches face challenges such as limited data and computational resources, especially with Transformers. TL addresses these issues by providing more robust PD diagnosis and better generalizability across languages. DAFE aims to improve the explainability and interpretability of results by examining the specific effects of deep features on both other DL approaches and more traditional machine learning (ML) methods. However, it often underperforms compared to E2E and TL approaches. Full article

This paper aims to present an approach to address the atom rearrangement problem by developing Convolutional Neural Network (CNN) models. These models predict the coordinates for atom movements while ensuring collision-free transitions and filling all vacancies in the target array. The process begins with designing a cost function for the assignment problem that incorporates constraints to prevent collisions and guarantee vacancy filling. We then build and train CNN models using datasets for three different grid sizes: 10×10,13×13,and21×21. Our models achieve high accuracy in predicting atom positions, with individual position accuracies of 99.63%, 98.93%, and 97.24%, respectively, for the aforementioned grid sizes. Despite limitations in training larger models due to hardware constraints, our approach demonstrates significant improvements in speed and accuracy. The final section of the paper presents detailed accuracy results and calculation times for each model, highlighting the potential of CNN-based methods in optimizing atom rearrangement processes. Full article

To enhance the detection of pavement-sealed cracks and ensure the long-term stability of pavement performance, a novel approach called the shuffle attention-based pavement-sealed crack detection is proposed. This method consists of three essential components: the feature extraction network, the detection head, and the Wise Intersection over Union loss function. Within both the feature extraction network and the detection head, the shuffle attention module is integrated to capture the high-dimensional semantic information of pavement-sealed cracks by combining spatial and channel attention in parallel. The two-way detection head with multi-scale feature fusion efficiently combines contextual information for pavement-sealed crack detection. Additionally, the Wise Intersection over Union loss function dynamically adjusts the gradient gain, enhancing the accuracy of bounding box fitting and coverage area. Experimental results highlight the superiority of our proposed method, with higher [email protected] (98.02%), Recall (0.9768), and F1-score (0.9680) values compared to the one-stage state-of-the-art methods, showcasing improvements of 0.81%, 1.8%, and 2.79%, respectively. Full article

Network function virtualization (NFV) allows the dynamic configuration of virtualized network functions to adapt services to complex and real-time network environments to improve network performance. The dynamic nature of physical networks creates significant challenges for virtual network function (VNF) migration and energy consumption, especially in edge–cloud continuum networks. This challenge can be addressed by predicting network traffic and proactively migrating VNFs using the predicted values. However, historical network traffic data are held by network service providers, and different network service providers are reluctant to share historical data due to privacy concerns; in addition, network resource providers that own the underlying networks are unable to effectively predict network traffic. To address this challenge, we apply a federated learning (FL) framework to enable network resource providers to no longer need historical network traffic data to be able to effectively predict network traffic. Further, to enable the predicted network traffic to lead to better migration effects, such as reducing the number of migrations, decreasing energy consumption, and increasing the request acceptance rate, we apply the predicted values of the network traffic to the network environment and feed the migration results of the network environment on the multiple factors described above to the neural network model. To obtain the migration results of the network environment, we analyzed and developed mathematical models for edge–cloud continuum networks with multiple network service providers. The effectiveness of our algorithm is evaluated through extensive simulations, and the results show a significant reduction in the number of migrated nodes and energy consumption, as well as an increase in the acceptance rate of the service function chain (SFC), compared with the commonly used scheme that uses only the difference between the predicted and actual traffic to define the loss function. Full article

The increasing sophistication of cyberattacks necessitates the development of advanced detection systems capable of accurately identifying and mitigating potential threats. This research addresses the critical challenge of cyberattack detection by employing a comprehensive approach that includes generating a realistic yet imbalanced dataset simulating various types of cyberattacks. Recognizing the inherent limitations posed by imbalanced data, we explored multiple data augmentation techniques to enhance the model’s learning effectiveness and ensure robust performance across different attack scenarios. Firstly, we constructed a detailed dataset reflecting real-world conditions of network intrusions by simulating a range of cyberattack types, ensuring it embodies the typical imbalances observed in genuine cybersecurity threats. Subsequently, we applied several data augmentation techniques, including SMOTE and ADASYN, to address the skew in class distribution, thereby providing a more balanced dataset for training supervised machine learning models. Our evaluation of these techniques across various models, such as Random Forests and Neural Networks, demonstrates significant improvements in detection capabilities. Moreover, the analysis also extends to the investigation of feature importance, providing critical insights into which attributes most significantly influence the predictive outcomes of the models. This not only enhances the interpretability of the models but also aids in refining feature engineering and selection processes to optimize performance. Full article

With the development of the power system, power users begin to use their own power supply in order to improve the power economy, but this also leads to the occurrence of the risk of self-provided power supply. The actual distribution network has few samples of power supply risk and it is difficult to identify the power supply risk by using conventional deep learning methods. In order to achieve high accuracy of self-provided power supply risk identification with small samples, this paper proposes a combination of transfer learning, convolutional block attention module (CBAM), and convolutional neural network (CNN) to identify the risk of self-provided power supply in an active distribution network. Firstly, in order to be able to further identify whether or not a risk will be caused based on completing the identification of the faulty line, we propose that it is necessary to identify whether or not the captive power supply on the faulty line is in operation. Second, in order to achieve high-precision identification and high-efficiency feature extraction, we propose to embed the CBAM into a CNN to form a CBAM-CNN model, so as to achieve high-efficiency feature extraction and high-precision risk identification. Finally, the use of transfer learning is proposed to solve the problem of low risk identification accuracy due to the small number of actual fault samples. Simulation experiments show that compared with other methods, the proposed method has the highest recognition accuracy and the best effect, and the risk recognition accuracy of active distribution network backup power is high in the case of fewer samples. Full article

Obesity recognition in adolescents is a growing concern. This study presents a deep learning-based obesity identification framework that integrates smartphone inertial measurements with deep learning models to address this issue. Utilizing data from accelerometers, gyroscopes, and rotation vectors collected via a mobile health application, we analyzed gait patterns for obesity indicators. Our framework employs three deep learning models: convolutional neural networks (CNNs), long-short-term memory network (LSTM), and a hybrid CNN–LSTM model. Trained on data from 138 subjects, including both normal and obese individuals, and tested on an additional 35 subjects, the hybrid model achieved the highest accuracy of 97%, followed by the LSTM model at 96.31% and the CNN model at 95.81%. Despite the promising outcomes, the study has limitations, such as a small sample and the exclusion of individuals with distorted gait. In future work, we aim to develop more generalized models that accommodate a broader range of gait patterns, including those with medical conditions. Full article

Traditional deep learning methods employing 2D images can only classify healthy and unhealthy seedlings; consequently, this study proposes a method by which to further classify healthy seedlings into primary seedlings and secondary seedlings and finally to differentiate three classes of seedling through a 3D point cloud for the detection of useful eggplant seedling transplants. Initially, RGB images of three types of substrate-cultivated eggplant seedlings (primary, secondary, and unhealthy) were collected, and healthy and unhealthy seedlings were classified using ResNet50, VGG16, and MobilNetV2. Subsequently, a 3D point cloud was generated for the three seedling types, and a series of filtering processes (fast Euclidean clustering, point cloud filtering, and voxel filtering) were employed to remove noise. Parameters (number of leaves, plant height, and stem diameter) extracted from the point cloud were found to be highly correlated with the manually measured values. The box plot shows that the primary and secondary seedlings were clearly differentiated for the extracted parameters. The point clouds of the three seedling types were ultimately classified directly using the 3D classification models PointNet++, dynamic graph convolutional neural network (DGCNN), and PointConv, in addition to the point cloud complementary operation for plants with missing leaves. The PointConv model demonstrated the best performance, with an average accuracy, precision, and recall of 95.83, 95.83, and 95.88%, respectively, and a model loss of 0.01. This method employs spatial feature information to analyse different seedling categories more effectively than two-dimensional (2D) image classification and three-dimensional (3D) feature extraction methods. However, there is a paucity of studies applying 3D classification methods to predict useful eggplant seedling transplants. Consequently, this method has the potential to identify different eggplant seedling types with high accuracy. Furthermore, it enables the quality inspection of seedlings during agricultural production. Full article

In the realm of precision space applications, improving the accuracy of orbit determination (OD) is a crucial and demanding task, primarily because of the presence of measurement noise. To address this issue, a novel machine learning method based on bidirectional long short-term memory (BiLSTM) is proposed in this research. In particular, the proposed method aims to improve the OD accuracy of Earth–Moon Libration orbits with angle-only measurements. The proposed BiLSTM network is designed to detect inaccurate measurements during an OD process, which is achieved by incorporating the least square method (LSM) as a basic estimation approach. The structure, inputs, and outputs of the modified BiLSTM network are meticulously crafted for the detection of inaccurate measurements. Following the detection of inaccurate measurements, a compensating strategy is devised to modify these detection results and thereby reduce their negative impact on OD accuracy. The modified measurements are then used to obtain a more accurate OD solution. The proposed method is applied to solve the OD problem of a 4:1 synodic resonant near-rectilinear halo orbit around the Earth–Moon L2 point. The training results reveal that the bidirectional network structure outperforms the regular unidirectional structures in terms of detection accuracy. Numerical simulations show that the proposed method can reduce the estimated error by approximately 10%. The proposed method holds significant potential for future missions in cislunar space. Full article

Obstacle avoidance in UAV swarms is crucial for ensuring the stability and safety of cluster flights. However, traditional methods of swarm obstacle avoidance often fail to meet the requirements of frequent spatiotemporal dynamic changes in UAV swarms, especially in complex environments such as forest firefighting, mine monitoring, and earthquake disaster relief. Consequently, the trained obstacle avoidance strategy differs from the expected or optimal obstacle avoidance scheme, leading to decision bias. To solve this problem, this paper proposes a method of UAV swarm obstacle avoidance decision making based on the end-edge-cloud collaboration model. In this method, the UAV swarm generates training data through environmental interaction. Sparse rewards are converted into dense rewards, considering the complex environmental state information and limited resources, and the actions of the UAVs are evaluated according to the reward values, to accurately assess the advantages and disadvantages of each agent’s actions. Finally, the training data and evaluation signals are utilized to optimize the parameters of the neural network through strategy-updating operations, aiming to improve the decision-making strategy. The experimental results demonstrate that the UAV swarm obstacle avoidance method proposed in this paper exhibits high obstacle avoidance efficiency, swarm stability, and completeness compared to other obstacle avoidance methods. Full article

One of the challenges in the drying process is decreasing the drying time while preserving the product quality. This work aimed to assess the impact of pulsed electric field (PEF) treatment with varying specific energy levels (15.2–26.8 kJ/kg) in conjunction with a microwave vacuum dryer (operating at energy levels of 100, 200 and 300 W) on the kinetics of drying apple slices (cv. Gravenstein). The findings demonstrated a notable reduction in the moisture ratio with the application of pulsed electric field treatment. Based on the findings, implementing PEF reduced the drying time from 4.2 to 31.4% compared to the untreated sample. Moreover, two mathematical models (viz. Page and Weibull) and two machine learning techniques (viz. artificial neural network and support vector regression) were used to predict the moisture ratio of the dried samples. Page’s and Weibull’s models predicted the moisture ratios with R² = 0.958 and 0.970, respectively. The optimal topology of machine learning to predict the moisture ratio was derived based on the influential parameters within the artificial neural network (i.e., training algorithm, transfer function and hidden layer neurons) and support vector regression (kernel function). The performance of the artificial neural network (R² = 0.998, RMSE = 0.038 and MAE = 0.024) surpassed that of support vector regression (R² = 0.994, RMSE = 0.012 and MAE = 0.009). Overall, the machine learning approach outperformed the mathematical models in terms of performance. Hence, machine learning can be used effectively for both predicting the moisture ratio and facilitating online monitoring and control of the drying processes. Lastly, the attributes of the dried apple slices, including color, mechanical properties and sensory analysis, were evaluated. Drying apple slices using PEF treatment and 100 W of microwave energy not only reduces drying time but also maintains the chemical properties such as the total phenolic content, total flavonoid content, antioxidant activity), vitamin C, color and sensory qualities of the product. Full article