The increased interest in geothermal energy is evident,along with the exploitation of traditional hydrothermal systems,in the growing research and projects developing around the reuse of already-drilled oil,gas,and ex...The increased interest in geothermal energy is evident,along with the exploitation of traditional hydrothermal systems,in the growing research and projects developing around the reuse of already-drilled oil,gas,and exploration wells.The Republic of Croatia has around 4000 wells,however,due to a long period since most of these wells were drilled and completed,there is uncertainty about how many are available for retrofitting as deep-borehole heat exchangers.Nevertheless,as hydrocarbon production decreases,it is expected that the number of wells available for the revitalization and exploitation of geothermal energy will increase.The revitalization of wells via deep-borehole heat exchangers involves installing a coaxial heat exchanger and circulating the working fluid in a closed system,during which heat is transferred from the surrounding rock medium to the circulating fluid.Since drilled wells are not of uniformdepth and are located in areas with different thermal rock properties and geothermal gradients,an analysis was conducted to determine available thermal energy as a function of well depth,geothermal gradient,and circulating fluid flow rate.Additionally,an economic analysis was performed to determine the benefits of retrofitting existing assets,such as drilled wells,compared to drilling new wells to obtain the same amount of thermal energy.展开更多
At present,energy consumption is one of the main bottlenecks in autonomous mobile robot development.To address the challenge of high energy consumption in path planning for autonomous mobile robots navigating unknown ...At present,energy consumption is one of the main bottlenecks in autonomous mobile robot development.To address the challenge of high energy consumption in path planning for autonomous mobile robots navigating unknown and complex environments,this paper proposes an Attention-Enhanced Dueling Deep Q-Network(ADDueling DQN),which integrates a multi-head attention mechanism and a prioritized experience replay strategy into a Dueling-DQN reinforcement learning framework.A multi-objective reward function,centered on energy efficiency,is designed to comprehensively consider path length,terrain slope,motion smoothness,and obstacle avoidance,enabling optimal low-energy trajectory generation in 3D space from the source.The incorporation of a multihead attention mechanism allows the model to dynamically focus on energy-critical state features—such as slope gradients and obstacle density—thereby significantly improving its ability to recognize and avoid energy-intensive paths.Additionally,the prioritized experience replay mechanism accelerates learning from key decision-making experiences,suppressing inefficient exploration and guiding the policy toward low-energy solutions more rapidly.The effectiveness of the proposed path planning algorithm is validated through simulation experiments conducted in multiple off-road scenarios.Results demonstrate that AD-Dueling DQN consistently achieves the lowest average energy consumption across all tested environments.Moreover,the proposed method exhibits faster convergence and greater training stability compared to baseline algorithms,highlighting its global optimization capability under energy-aware objectives in complex terrains.This study offers an efficient and scalable intelligent control strategy for the development of energy-conscious autonomous navigation systems.展开更多
Unmanned Aerial Vehicle(UAV)plays a prominent role in various fields,and autonomous navigation is a crucial component of UAV intelligence.Deep Reinforcement Learning(DRL)has expanded the research avenues for addressin...Unmanned Aerial Vehicle(UAV)plays a prominent role in various fields,and autonomous navigation is a crucial component of UAV intelligence.Deep Reinforcement Learning(DRL)has expanded the research avenues for addressing challenges in autonomous navigation.Nonetheless,challenges persist,including getting stuck in local optima,consuming excessive computations during action space exploration,and neglecting deterministic experience.This paper proposes a noise-driven enhancement strategy.In accordance with the overall learning phases,a global noise control method is designed,while a differentiated local noise control method is developed by analyzing the exploration demands of four typical situations encountered by UAV during navigation.Both methods are integrated into a dual-model for noise control to regulate action space exploration.Furthermore,noise dual experience replay buffers are designed to optimize the rational utilization of both deterministic and noisy experience.In uncertain environments,based on the Twin Delay Deep Deterministic Policy Gradient(TD3)algorithm with Long Short-Term Memory(LSTM)network and Priority Experience Replay(PER),a Noise-Driven Enhancement Priority Memory TD3(NDE-PMTD3)is developed.We established a simulation environment to compare different algorithms,and the performance of the algorithms is analyzed in various scenarios.The training results indicate that the proposed algorithm accelerates the convergence speed and enhances the convergence stability.In test experiments,the proposed algorithm successfully and efficiently performs autonomous navigation tasks in diverse environments,demonstrating superior generalization results.展开更多
With the continuous development of artificial intelligence and computer vision technology,numerous deep learning-based lane line detection methods have emerged.DeepLabv3+,as a classic semantic segmentation model,has f...With the continuous development of artificial intelligence and computer vision technology,numerous deep learning-based lane line detection methods have emerged.DeepLabv3+,as a classic semantic segmentation model,has found widespread application in the field of lane line detection.However,the accuracy of lane line segmentation is often compromised by factors such as changes in lighting conditions,occlusions,and wear and tear on the lane lines.Additionally,DeepLabv3+suffers from high memory consumption and challenges in deployment on embedded platforms.To address these issues,this paper proposes a lane line detection method for complex road scenes based on DeepLabv3+and MobileNetV4(MNv4).First,the lightweight MNv4 is adopted as the backbone network,and the standard convolutions in ASPP are replaced with depthwise separable convolutions.Second,a polarization attention mechanism is introduced after the ASPP module to enhance the model’s generalization capability.Finally,the Simple Linear Iterative Clustering(SLIC)superpixel segmentation algorithmis employed to preserve lane line edge information.MNv4-DeepLabv3+was tested on the TuSimple and CULane datasets.On the TuSimple dataset,theMean Intersection over Union(MIoU)and Mean Pixel Accuracy(mPA)improved by 1.01%and 7.49%,respectively.On the CULane dataset,MIoU andmPA increased by 3.33%and 7.74%,respectively.Thenumber of parameters decreased from 54.84 to 3.19 M.Experimental results demonstrate that MNv4-DeepLabv3+significantly optimizes model parameter count and enhances segmentation accuracy.展开更多
Underwater pipeline inspection plays a vital role in the proactive maintenance and management of critical marine infrastructure and subaquatic systems.However,the inspection of underwater pipelines presents a challeng...Underwater pipeline inspection plays a vital role in the proactive maintenance and management of critical marine infrastructure and subaquatic systems.However,the inspection of underwater pipelines presents a challenge due to factors such as light scattering,absorption,restricted visibility,and ambient noise.The advancement of deep learning has introduced powerful techniques for processing large amounts of unstructured and imperfect data collected from underwater environments.This study evaluated the efficacy of the You Only Look Once(YOLO)algorithm,a real-time object detection and localization model based on convolutional neural networks,in identifying and classifying various types of pipeline defects in underwater settings.YOLOv8,the latest evolution in the YOLO family,integrates advanced capabilities,such as anchor-free detection,a cross-stage partial network backbone for efficient feature extraction,and a feature pyramid network+path aggregation network neck for robust multi-scale object detection,which make it particularly well-suited for complex underwater environments.Due to the lack of suitable open-access datasets for underwater pipeline defects,a custom dataset was captured using a remotely operated vehicle in a controlled environment.This application has the following assets available for use.Extensive experimentation demonstrated that YOLOv8 X-Large consistently outperformed other models in terms of pipe defect detection and classification and achieved a strong balance between precision and recall in identifying pipeline cracks,rust,corners,defective welds,flanges,tapes,and holes.This research establishes the baseline performance of YOLOv8 for underwater defect detection and showcases its potential to enhance the reliability and efficiency of pipeline inspection tasks in challenging underwater environments.展开更多
AlGaN-based deep-ultraviolet(DUV)laser diodes(LDs)face performance challenges due to elec-tron leakage and poor hole injection which is often worsened by polarization effects from conventional elec-tron blocking layer...AlGaN-based deep-ultraviolet(DUV)laser diodes(LDs)face performance challenges due to elec-tron leakage and poor hole injection which is often worsened by polarization effects from conventional elec-tron blocking layers(EBLs).To overcome these limitations,we propose an EBL-free DUV LD design incor-porating a 1-nm undoped Al_(0.8)Ga_(0.2)N thin strip layer after the last quantum barrier.Using PICS3D simula-tions,we evaluate the optical and electrical characteristics.Results show a significant increase in effective electron barrier height(from 158.2 meV to 420.7 meV)and a reduction in hole barrier height(from 149.2 meV to 62.8 meV),which enhance hole injection and reduce electron leakage.The optimized structure(LD3)achieves a 14%increase in output power,improved slope efficiency(1.85 W/A),and lower threshold current.This design also reduces the quantum confined Stark effect and forms dual hole accumulation regions,im-proving recombination efficiency.展开更多
Artificial Intelligence(AI)in healthcare enables predicting diabetes using data-driven methods instead of the traditional ways of screening the disease,which include hemoglobin A1c(HbA1c),oral glucose tolerance test(O...Artificial Intelligence(AI)in healthcare enables predicting diabetes using data-driven methods instead of the traditional ways of screening the disease,which include hemoglobin A1c(HbA1c),oral glucose tolerance test(OGTT),and fasting plasma glucose(FPG)screening techniques,which are invasive and limited in scale.Machine learning(ML)and deep neural network(DNN)models that use large datasets to learn the complex,nonlinear feature interactions,but the conventional ML algorithms are data sensitive and often show unstable predictive accuracy.Conversely,DNN models are more robust,though the ability to reach a high accuracy rate consistently on heterogeneous datasets is still an open challenge.For predicting diabetes,this work proposed a hybrid DNN approach by integrating a bidirectional long short-term memory(BiLSTM)network with a bidirectional gated recurrent unit(BiGRU).A robust DL model,developed by combining various datasets with weighted coefficients,dense operations in the connection of deep layers,and the output aggregation using batch normalization and dropout functions to avoid overfitting.The goal of this hybrid model is better generalization and consistency among various datasets,which facilitates the effective management and early intervention.The proposed DNN model exhibits an excellent predictive performance as compared to the state-of-the-art and baseline ML and DNN models for diabetes prediction tasks.The robust performance indicates the possible usefulness of DL-based models in the development of disease prediction in healthcare and other areas that demand high-quality analytics.展开更多
Accurate streamflow prediction is essential for flood warning,reservoir operation,irrigation scheduling,hydropower planning,and sustainable water management,yet remains challenging due to the complexity of hydrologica...Accurate streamflow prediction is essential for flood warning,reservoir operation,irrigation scheduling,hydropower planning,and sustainable water management,yet remains challenging due to the complexity of hydrological processes.Although data-driven models often outperform conventional physics-based hydrological modelling approaches,their real-world deployment is limited by cost,infrastructure demands,and the interdisciplinary expertise required.To bridge this gap,this study developed QPred,a regional,lightweight,cost-effective,web-delivered application for daily streamflow forecasting.The study executed an end-to-end workflow,from field data acquisition to accessible web-based deployment for on-demand forecasting.High-resolution rainfall data were recorded with tippingbucket gauges and loggers,while river water depth in the Aglar and Paligaad watersheds was converted to discharge using site-specific rating curves,resulting in a daily dataset of precipitation,river water level and discharge.Four DL architectures were trained,including vanilla Long Short-Term Memory(LSTM),stacked LSTM,bidirectional LSTM,and Gated Recurrent Unit(GRU),and evaluated using Nash-Sutcliffe Efficiency(NSE),Coefficient of Determination(R2),Root-Mean-Square-Error-Standard-Deviation Ratio(RSR),and Percentage Bias(PBIAS)metrics.Performance was watershed-specific,as the vanilla LSTM demonstrated the best generalisation for the Aglar watershed(R2=0.88,NSE=0.82,RMSE=0.12 during validation),while the GRU achieved the highest validation accuracy in Paligaad(R2=0.88,NSE=0.88,RMSE=0.49).All models achieved satisfactory to excellent performance during calibration(R2>0.91,NSE>0.91 for both watersheds),demonstrating strong capability to capture streamflow dynamics.The highest performing models were selected and embedded into the QPred application.QPred was developed as a lightweight web pipeline,utilising Google Colab as the primary execution environment,Flask as the backend inference framework,Google Drive for artefact storage,andNgrok for secureHTTPS tunnelling.Auser-friendly front end utilises range sliders(bounded by observed minima and maxima)to gather inputs and provides discharge data along with metadata,thereby enhancing transparency.This work demonstrates that accurate,context-aware deep learningmodels can be delivered through low-cost,web-based platforms,providing a reproducible and scalable pipeline for hydrological applications in other watersheds and for practitioners.展开更多
This paper proposes a deep learning-based 3D LiDAR perception framework designed for applications such as autonomous robots and vehicles.To address the high dependency on large-scale annotated data—an inherent limita...This paper proposes a deep learning-based 3D LiDAR perception framework designed for applications such as autonomous robots and vehicles.To address the high dependency on large-scale annotated data—an inherent limitation of deep learning models—this study introduces a hybrid perception architecture that incorporates expertdriven LiDAR processing techniques into the deep neural network.Traditional 3DLiDAR processingmethods typically remove ground planes and apply distance-or density-based clustering for object detection.In this work,such expert knowledge is encoded as feature-level inputs and fused with the deep network,therebymitigating the data dependency issue of conventional learning-based approaches.Specifically,the proposedmethod combines two expert algorithms—Patchwork++for ground segmentation and DBSCAN for clustering—with a PointPillars-based LiDAR detection network.We design four hybrid versions of the network depending on the stage and method of integrating expert features into the feature map of the deep model.Among these,Version 4 incorporates a modified neck structure in PointPillars and introduces a new Cluster 2D Pseudo-Map Branch that utilizes cluster-level pseudo-images generated from Patchwork++and DBSCAN.This version achieved a+3.88%improvement mean Average Precision(mAP)compared to the baseline PointPillars.The results demonstrate that embedding expert-based perception logic into deep neural architectures can effectively enhance performance and reduce dependency on extensive training datasets,offering a promising direction for robust 3D LiDAR object detection in real-world scenarios.展开更多
To achieve the potential performance gain of massive multiple-input multiple-output(MIMO)systems,base stations(BS)require downlink channel state information(CSI)fed back by users to execute beamforming design,especial...To achieve the potential performance gain of massive multiple-input multiple-output(MIMO)systems,base stations(BS)require downlink channel state information(CSI)fed back by users to execute beamforming design,especially in the frequency division duplex(FDD)systems.However,due to the enormous number of antennas in massive MIMO systems,the feedback overhead of downlink CSI acquisition is extremely large.To address this issue,deep learning(DL)techniques have been introduced to de velop high-accuracy feedback strategies under limited backhaul constraints.In this paper,we provide an overview of DL-based CSI compression and feedback approaches in massive MIMO systems.Specifically,we introduce the conventional CSI compression and feedback schemes and the existing problems.Besides,we elaborate on various DL techniques employed in CSI compression from the perspective of network architecture and analyze the advantages of different techniques.We also enumerate the applications of DL-based methods for solving practical challenges in CSI compression and feedback.In addition,we brief the remaining issues in deep CSI compression and indicate potential directions in future wireless networks.展开更多
Wearable sensors integrated with deep learning techniques have the potential to revolutionize seamless human-machine interfaces for real-time health monitoring,clinical diagnosis,and robotic applications.Nevertheless,...Wearable sensors integrated with deep learning techniques have the potential to revolutionize seamless human-machine interfaces for real-time health monitoring,clinical diagnosis,and robotic applications.Nevertheless,it remains a critical challenge to simultaneously achieve desirable mechanical and electrical performance along with biocompatibility,adhesion,self-healing,and environmental robustness with excellent sensing metrics.Herein,we report a multifunctional,anti-freezing,selfadhesive,and self-healable organogel pressure sensor composed of cobalt nanoparticle encapsulated nitrogen-doped carbon nanotubes(CoN CNT)embedded in a polyvinyl alcohol-gelatin(PVA/GLE)matrix.Fabricated using a binary solvent system of water and ethylene glycol(EG),the CoN CNT/PVA/GLE organogel exhibits excellent flexibility,biocompatibility,and temperature tolerance with remarkable environmental stability.Electrochemical impedance spectroscopy confirms near-stable performance across a broad humidity range(40%-95%RH).Freeze-tolerant conductivity under sub-zero conditions(-20℃)is attributed to the synergistic role of CoN CNT and EG,preserving mobility and network integrity.The Co N CNT/PVA/GLE organogel sensor exhibits high sensitivity of 5.75 k Pa^(-1)in the detection range from 0 to 20 k Pa,ideal for subtle biomechanical motion detection.A smart human-machine interface for English letter recognition using deep learning achieved 98%accuracy.The organogel sensor utility was extended to detect human gestures like finger bending,wrist motion,and throat vibration during speech.展开更多
The solar cycle(SC),a phenomenon caused by the quasi-periodic regular activities in the Sun,occurs approximately every 11 years.Intense solar activity can disrupt the Earth’s ionosphere,affecting communication and na...The solar cycle(SC),a phenomenon caused by the quasi-periodic regular activities in the Sun,occurs approximately every 11 years.Intense solar activity can disrupt the Earth’s ionosphere,affecting communication and navigation systems.Consequently,accurately predicting the intensity of the SC holds great significance,but predicting the SC involves a long-term time series,and many existing time series forecasting methods have fallen short in terms of accuracy and efficiency.The Time-series Dense Encoder model is a deep learning solution tailored for long time series prediction.Based on a multi-layer perceptron structure,it outperforms the best previously existing models in accuracy,while being efficiently trainable on general datasets.We propose a method based on this model for SC forecasting.Using a trained model,we predict the test set from SC 19 to SC 25 with an average mean absolute percentage error of 32.02,root mean square error of 30.3,mean absolute error of 23.32,and R^(2)(coefficient of determination)of 0.76,outperforming other deep learning models in terms of accuracy and training efficiency on sunspot number datasets.Subsequently,we use it to predict the peaks of SC 25 and SC 26.For SC 25,the peak time has ended,but a stronger peak is predicted for SC 26,of 199.3,within a range of 170.8-221.9,projected to occur during April 2034.展开更多
This study aimed to integrate Monte Carlo(MC)simulation with deep learning(DL)-based denoising techniques to achieve fast and accurate prediction of high-quality electronic portal imaging device(EPID)transmission dose...This study aimed to integrate Monte Carlo(MC)simulation with deep learning(DL)-based denoising techniques to achieve fast and accurate prediction of high-quality electronic portal imaging device(EPID)transmission dose(TD)for patientspecific quality assurance(PSQA).A total of 100 lung cases were used to obtain the noisy EPID TD by the ARCHER MC code under four kinds of particle numbers(1×10^(6),1×10^(7),1×10^(8)and 1×10^(9)),and the original EPID TD was denoised by the SUNet neural network.The denoised EPID TD was assessed both qualitatively and quantitatively using the structural similarity(SSIM),peak signal-to-noise ratio(PSNR),and gamma passing rate(GPR)with respect to 1×10^(9)as a reference.The computation times for both the MC simulation and DL-based denoising were recorded.As the number of particles increased,both the quality of the noisy EPID TD and computation time increased significantly(1×10^(6):1.12 s,1×10^(7):1.72 s,1×10^(8):8.62 s,and 1×10^(9):73.89 s).In contrast,the DL-based denoising time remained at 0.13-0.16 s.The denoised EPID TD shows a smoother visual appearance and profile curves,but differences between 1×10^(6)and 1×10^(9)still remain.SSIM improves from 0.61 to 0.95 for 1×10^(6),0.70 to 0.96 for 1×10^(7),and 0.90 to 0.97 for 1×10^(8).PSNR increases by>20%for 1×10^(6)and 1×10^(7),and>10%for 1×10^(8).GPR improves from 48.47%to 89.10%for 1×10^(6),61.04%to 94.35%for 1×10^(7),and 91.88%to 99.55%for 1×10^(8).The method that combines MC simulation with DL-based denoising for EPID TD generation can accelerate TD prediction and maintain high accuracy,offering a promising solution for efficient PSQA.展开更多
Green solvent pretreatment of biomass represents a promising ap-proach for enhancing the econom-ic value of lignocellulosic deriva-tives.In this study,corncob biomass was treated with a diol-based deep eutectic solven...Green solvent pretreatment of biomass represents a promising ap-proach for enhancing the econom-ic value of lignocellulosic deriva-tives.In this study,corncob biomass was treated with a diol-based deep eutectic solvent(DES)under mild conditions,facilitating efficient cellulose separation.The extracted cellulose was subsequently used to fabricate cellulose hydrogels in an aqueous zinc chloride solution.The resulting hydrogel exhibited a“water-in-salt”effect due to the high concentration of ZnCl_(2).Leveraging the antifreeze properties of sorbitol,the system demon-strated outstanding low-temperature electrochemical performance,including a broad operat-ing voltage window and an ionic conductivity of 38.4 mS·cm^(-1)at-20℃.At 20℃,the de-vice achieved an energy density of 206 Wh·kg^(-1)and a power density of 2701.05 W·kg^(-1)at a current density of 1 A·g^(-1).Moreover,the flexible zinc-ion hybrid supercapacitor(ZHSC)maintained 89%of its capacitance and nearly 100%Coulombic efficiency after 5500 cycles at 20℃.This work not only advances the development of zinc-ion energy storage devices but al-so establishes a new paradigm for the green and direct utilization of biomass-derived materi-als.展开更多
Carnauba wax-structured soybean oil oleogels were developed as alternative frying media for proso millet rice cake processing.Optimal practical properties were identified in the 8%wax concentration through evaluations...Carnauba wax-structured soybean oil oleogels were developed as alternative frying media for proso millet rice cake processing.Optimal practical properties were identified in the 8%wax concentration through evaluations conducted via polarized light microscopy,oil holding capacity,freeze-thaw stability,and rheological analysis.Oleogel-fried cakes were found to exhibit 6.69%–19.98%lower oil absorption compared to soybean oil-fried samples during 0–18 h frying.The corresponding low-field NMR results demonstrated that the percentage of the relaxation peak area attributed to oil content in oleogel-fried proso millet rice cakes was also reduced.The lipid oxidation rates were significantly reduced in oleogels(P<0.05),with total polar compounds(9.89%–22.81%)and acid values(4.21%–24.37%)decreased while thermal stability was maintained.Saturated fatty acid levels in oleogel were reduced by 3.18%–7.26%after frying for 12–18 h.Texture properties were slightly affected whereas color attributes were significantly improved.Enhanced texture,flavor,hardness,and mouthfeel in oleogel-fried products were confirmed through sensory evaluation.Volatile components,particularly ketones and esters were significantly increased in rice cakes fried with oleogel.These results might offer novel insights for advancement of future low-oil frying technologies and expand the application of oleogels.展开更多
文摘The increased interest in geothermal energy is evident,along with the exploitation of traditional hydrothermal systems,in the growing research and projects developing around the reuse of already-drilled oil,gas,and exploration wells.The Republic of Croatia has around 4000 wells,however,due to a long period since most of these wells were drilled and completed,there is uncertainty about how many are available for retrofitting as deep-borehole heat exchangers.Nevertheless,as hydrocarbon production decreases,it is expected that the number of wells available for the revitalization and exploitation of geothermal energy will increase.The revitalization of wells via deep-borehole heat exchangers involves installing a coaxial heat exchanger and circulating the working fluid in a closed system,during which heat is transferred from the surrounding rock medium to the circulating fluid.Since drilled wells are not of uniformdepth and are located in areas with different thermal rock properties and geothermal gradients,an analysis was conducted to determine available thermal energy as a function of well depth,geothermal gradient,and circulating fluid flow rate.Additionally,an economic analysis was performed to determine the benefits of retrofitting existing assets,such as drilled wells,compared to drilling new wells to obtain the same amount of thermal energy.
文摘At present,energy consumption is one of the main bottlenecks in autonomous mobile robot development.To address the challenge of high energy consumption in path planning for autonomous mobile robots navigating unknown and complex environments,this paper proposes an Attention-Enhanced Dueling Deep Q-Network(ADDueling DQN),which integrates a multi-head attention mechanism and a prioritized experience replay strategy into a Dueling-DQN reinforcement learning framework.A multi-objective reward function,centered on energy efficiency,is designed to comprehensively consider path length,terrain slope,motion smoothness,and obstacle avoidance,enabling optimal low-energy trajectory generation in 3D space from the source.The incorporation of a multihead attention mechanism allows the model to dynamically focus on energy-critical state features—such as slope gradients and obstacle density—thereby significantly improving its ability to recognize and avoid energy-intensive paths.Additionally,the prioritized experience replay mechanism accelerates learning from key decision-making experiences,suppressing inefficient exploration and guiding the policy toward low-energy solutions more rapidly.The effectiveness of the proposed path planning algorithm is validated through simulation experiments conducted in multiple off-road scenarios.Results demonstrate that AD-Dueling DQN consistently achieves the lowest average energy consumption across all tested environments.Moreover,the proposed method exhibits faster convergence and greater training stability compared to baseline algorithms,highlighting its global optimization capability under energy-aware objectives in complex terrains.This study offers an efficient and scalable intelligent control strategy for the development of energy-conscious autonomous navigation systems.
基金the Collaborative Innovation Project of Shanghai,China for the financial support。
文摘Unmanned Aerial Vehicle(UAV)plays a prominent role in various fields,and autonomous navigation is a crucial component of UAV intelligence.Deep Reinforcement Learning(DRL)has expanded the research avenues for addressing challenges in autonomous navigation.Nonetheless,challenges persist,including getting stuck in local optima,consuming excessive computations during action space exploration,and neglecting deterministic experience.This paper proposes a noise-driven enhancement strategy.In accordance with the overall learning phases,a global noise control method is designed,while a differentiated local noise control method is developed by analyzing the exploration demands of four typical situations encountered by UAV during navigation.Both methods are integrated into a dual-model for noise control to regulate action space exploration.Furthermore,noise dual experience replay buffers are designed to optimize the rational utilization of both deterministic and noisy experience.In uncertain environments,based on the Twin Delay Deep Deterministic Policy Gradient(TD3)algorithm with Long Short-Term Memory(LSTM)network and Priority Experience Replay(PER),a Noise-Driven Enhancement Priority Memory TD3(NDE-PMTD3)is developed.We established a simulation environment to compare different algorithms,and the performance of the algorithms is analyzed in various scenarios.The training results indicate that the proposed algorithm accelerates the convergence speed and enhances the convergence stability.In test experiments,the proposed algorithm successfully and efficiently performs autonomous navigation tasks in diverse environments,demonstrating superior generalization results.
文摘With the continuous development of artificial intelligence and computer vision technology,numerous deep learning-based lane line detection methods have emerged.DeepLabv3+,as a classic semantic segmentation model,has found widespread application in the field of lane line detection.However,the accuracy of lane line segmentation is often compromised by factors such as changes in lighting conditions,occlusions,and wear and tear on the lane lines.Additionally,DeepLabv3+suffers from high memory consumption and challenges in deployment on embedded platforms.To address these issues,this paper proposes a lane line detection method for complex road scenes based on DeepLabv3+and MobileNetV4(MNv4).First,the lightweight MNv4 is adopted as the backbone network,and the standard convolutions in ASPP are replaced with depthwise separable convolutions.Second,a polarization attention mechanism is introduced after the ASPP module to enhance the model’s generalization capability.Finally,the Simple Linear Iterative Clustering(SLIC)superpixel segmentation algorithmis employed to preserve lane line edge information.MNv4-DeepLabv3+was tested on the TuSimple and CULane datasets.On the TuSimple dataset,theMean Intersection over Union(MIoU)and Mean Pixel Accuracy(mPA)improved by 1.01%and 7.49%,respectively.On the CULane dataset,MIoU andmPA increased by 3.33%and 7.74%,respectively.Thenumber of parameters decreased from 54.84 to 3.19 M.Experimental results demonstrate that MNv4-DeepLabv3+significantly optimizes model parameter count and enhances segmentation accuracy.
文摘Underwater pipeline inspection plays a vital role in the proactive maintenance and management of critical marine infrastructure and subaquatic systems.However,the inspection of underwater pipelines presents a challenge due to factors such as light scattering,absorption,restricted visibility,and ambient noise.The advancement of deep learning has introduced powerful techniques for processing large amounts of unstructured and imperfect data collected from underwater environments.This study evaluated the efficacy of the You Only Look Once(YOLO)algorithm,a real-time object detection and localization model based on convolutional neural networks,in identifying and classifying various types of pipeline defects in underwater settings.YOLOv8,the latest evolution in the YOLO family,integrates advanced capabilities,such as anchor-free detection,a cross-stage partial network backbone for efficient feature extraction,and a feature pyramid network+path aggregation network neck for robust multi-scale object detection,which make it particularly well-suited for complex underwater environments.Due to the lack of suitable open-access datasets for underwater pipeline defects,a custom dataset was captured using a remotely operated vehicle in a controlled environment.This application has the following assets available for use.Extensive experimentation demonstrated that YOLOv8 X-Large consistently outperformed other models in terms of pipe defect detection and classification and achieved a strong balance between precision and recall in identifying pipeline cracks,rust,corners,defective welds,flanges,tapes,and holes.This research establishes the baseline performance of YOLOv8 for underwater defect detection and showcases its potential to enhance the reliability and efficiency of pipeline inspection tasks in challenging underwater environments.
文摘AlGaN-based deep-ultraviolet(DUV)laser diodes(LDs)face performance challenges due to elec-tron leakage and poor hole injection which is often worsened by polarization effects from conventional elec-tron blocking layers(EBLs).To overcome these limitations,we propose an EBL-free DUV LD design incor-porating a 1-nm undoped Al_(0.8)Ga_(0.2)N thin strip layer after the last quantum barrier.Using PICS3D simula-tions,we evaluate the optical and electrical characteristics.Results show a significant increase in effective electron barrier height(from 158.2 meV to 420.7 meV)and a reduction in hole barrier height(from 149.2 meV to 62.8 meV),which enhance hole injection and reduce electron leakage.The optimized structure(LD3)achieves a 14%increase in output power,improved slope efficiency(1.85 W/A),and lower threshold current.This design also reduces the quantum confined Stark effect and forms dual hole accumulation regions,im-proving recombination efficiency.
基金supported by the School of Digital Science,Universiti Brunei Darussalam,Brunei.
文摘Artificial Intelligence(AI)in healthcare enables predicting diabetes using data-driven methods instead of the traditional ways of screening the disease,which include hemoglobin A1c(HbA1c),oral glucose tolerance test(OGTT),and fasting plasma glucose(FPG)screening techniques,which are invasive and limited in scale.Machine learning(ML)and deep neural network(DNN)models that use large datasets to learn the complex,nonlinear feature interactions,but the conventional ML algorithms are data sensitive and often show unstable predictive accuracy.Conversely,DNN models are more robust,though the ability to reach a high accuracy rate consistently on heterogeneous datasets is still an open challenge.For predicting diabetes,this work proposed a hybrid DNN approach by integrating a bidirectional long short-term memory(BiLSTM)network with a bidirectional gated recurrent unit(BiGRU).A robust DL model,developed by combining various datasets with weighted coefficients,dense operations in the connection of deep layers,and the output aggregation using batch normalization and dropout functions to avoid overfitting.The goal of this hybrid model is better generalization and consistency among various datasets,which facilitates the effective management and early intervention.The proposed DNN model exhibits an excellent predictive performance as compared to the state-of-the-art and baseline ML and DNN models for diabetes prediction tasks.The robust performance indicates the possible usefulness of DL-based models in the development of disease prediction in healthcare and other areas that demand high-quality analytics.
文摘Accurate streamflow prediction is essential for flood warning,reservoir operation,irrigation scheduling,hydropower planning,and sustainable water management,yet remains challenging due to the complexity of hydrological processes.Although data-driven models often outperform conventional physics-based hydrological modelling approaches,their real-world deployment is limited by cost,infrastructure demands,and the interdisciplinary expertise required.To bridge this gap,this study developed QPred,a regional,lightweight,cost-effective,web-delivered application for daily streamflow forecasting.The study executed an end-to-end workflow,from field data acquisition to accessible web-based deployment for on-demand forecasting.High-resolution rainfall data were recorded with tippingbucket gauges and loggers,while river water depth in the Aglar and Paligaad watersheds was converted to discharge using site-specific rating curves,resulting in a daily dataset of precipitation,river water level and discharge.Four DL architectures were trained,including vanilla Long Short-Term Memory(LSTM),stacked LSTM,bidirectional LSTM,and Gated Recurrent Unit(GRU),and evaluated using Nash-Sutcliffe Efficiency(NSE),Coefficient of Determination(R2),Root-Mean-Square-Error-Standard-Deviation Ratio(RSR),and Percentage Bias(PBIAS)metrics.Performance was watershed-specific,as the vanilla LSTM demonstrated the best generalisation for the Aglar watershed(R2=0.88,NSE=0.82,RMSE=0.12 during validation),while the GRU achieved the highest validation accuracy in Paligaad(R2=0.88,NSE=0.88,RMSE=0.49).All models achieved satisfactory to excellent performance during calibration(R2>0.91,NSE>0.91 for both watersheds),demonstrating strong capability to capture streamflow dynamics.The highest performing models were selected and embedded into the QPred application.QPred was developed as a lightweight web pipeline,utilising Google Colab as the primary execution environment,Flask as the backend inference framework,Google Drive for artefact storage,andNgrok for secureHTTPS tunnelling.Auser-friendly front end utilises range sliders(bounded by observed minima and maxima)to gather inputs and provides discharge data along with metadata,thereby enhancing transparency.This work demonstrates that accurate,context-aware deep learningmodels can be delivered through low-cost,web-based platforms,providing a reproducible and scalable pipeline for hydrological applications in other watersheds and for practitioners.
基金supported by Basic Science Research Program through the National Research Foundation of Korea(NRF)funded by the Ministry of Education(RS-2023-00245084)by Korea Institute for Advancement of Technology(KIAT)grant funded by the Korea Government(MOTIE)(RS-2024-00415938,HRD Program for Industrial Innovation)and Soonchunhyang University.
文摘This paper proposes a deep learning-based 3D LiDAR perception framework designed for applications such as autonomous robots and vehicles.To address the high dependency on large-scale annotated data—an inherent limitation of deep learning models—this study introduces a hybrid perception architecture that incorporates expertdriven LiDAR processing techniques into the deep neural network.Traditional 3DLiDAR processingmethods typically remove ground planes and apply distance-or density-based clustering for object detection.In this work,such expert knowledge is encoded as feature-level inputs and fused with the deep network,therebymitigating the data dependency issue of conventional learning-based approaches.Specifically,the proposedmethod combines two expert algorithms—Patchwork++for ground segmentation and DBSCAN for clustering—with a PointPillars-based LiDAR detection network.We design four hybrid versions of the network depending on the stage and method of integrating expert features into the feature map of the deep model.Among these,Version 4 incorporates a modified neck structure in PointPillars and introduces a new Cluster 2D Pseudo-Map Branch that utilizes cluster-level pseudo-images generated from Patchwork++and DBSCAN.This version achieved a+3.88%improvement mean Average Precision(mAP)compared to the baseline PointPillars.The results demonstrate that embedding expert-based perception logic into deep neural architectures can effectively enhance performance and reduce dependency on extensive training datasets,offering a promising direction for robust 3D LiDAR object detection in real-world scenarios.
基金supported by ZTE Industry-University-Institute Cooperation Funds under Grant No.IA20240319003the NSFC under Grant No.62571112。
文摘To achieve the potential performance gain of massive multiple-input multiple-output(MIMO)systems,base stations(BS)require downlink channel state information(CSI)fed back by users to execute beamforming design,especially in the frequency division duplex(FDD)systems.However,due to the enormous number of antennas in massive MIMO systems,the feedback overhead of downlink CSI acquisition is extremely large.To address this issue,deep learning(DL)techniques have been introduced to de velop high-accuracy feedback strategies under limited backhaul constraints.In this paper,we provide an overview of DL-based CSI compression and feedback approaches in massive MIMO systems.Specifically,we introduce the conventional CSI compression and feedback schemes and the existing problems.Besides,we elaborate on various DL techniques employed in CSI compression from the perspective of network architecture and analyze the advantages of different techniques.We also enumerate the applications of DL-based methods for solving practical challenges in CSI compression and feedback.In addition,we brief the remaining issues in deep CSI compression and indicate potential directions in future wireless networks.
基金supported by the Basic Science Research Program(2023R1A2C3004336,RS-202300243807)&Regional Leading Research Center(RS-202400405278)through the National Research Foundation of Korea(NRF)grant funded by the Korea Government(MSIT)。
文摘Wearable sensors integrated with deep learning techniques have the potential to revolutionize seamless human-machine interfaces for real-time health monitoring,clinical diagnosis,and robotic applications.Nevertheless,it remains a critical challenge to simultaneously achieve desirable mechanical and electrical performance along with biocompatibility,adhesion,self-healing,and environmental robustness with excellent sensing metrics.Herein,we report a multifunctional,anti-freezing,selfadhesive,and self-healable organogel pressure sensor composed of cobalt nanoparticle encapsulated nitrogen-doped carbon nanotubes(CoN CNT)embedded in a polyvinyl alcohol-gelatin(PVA/GLE)matrix.Fabricated using a binary solvent system of water and ethylene glycol(EG),the CoN CNT/PVA/GLE organogel exhibits excellent flexibility,biocompatibility,and temperature tolerance with remarkable environmental stability.Electrochemical impedance spectroscopy confirms near-stable performance across a broad humidity range(40%-95%RH).Freeze-tolerant conductivity under sub-zero conditions(-20℃)is attributed to the synergistic role of CoN CNT and EG,preserving mobility and network integrity.The Co N CNT/PVA/GLE organogel sensor exhibits high sensitivity of 5.75 k Pa^(-1)in the detection range from 0 to 20 k Pa,ideal for subtle biomechanical motion detection.A smart human-machine interface for English letter recognition using deep learning achieved 98%accuracy.The organogel sensor utility was extended to detect human gestures like finger bending,wrist motion,and throat vibration during speech.
基金supported by the Academic Research Projects of Beijing Union University(ZK20202204)the National Natural Science Foundation of China(12250005,12073040,12273059,11973056,12003051,11573037,12073041,11427901,11572005,11611530679 and 12473052)+1 种基金the Strategic Priority Research Program of the China Academy of Sciences(XDB0560000,XDA15052200,XDB09040200,XDA15010700,XDB0560301,and XDA15320102)the Chinese Meridian Project(CMP).
文摘The solar cycle(SC),a phenomenon caused by the quasi-periodic regular activities in the Sun,occurs approximately every 11 years.Intense solar activity can disrupt the Earth’s ionosphere,affecting communication and navigation systems.Consequently,accurately predicting the intensity of the SC holds great significance,but predicting the SC involves a long-term time series,and many existing time series forecasting methods have fallen short in terms of accuracy and efficiency.The Time-series Dense Encoder model is a deep learning solution tailored for long time series prediction.Based on a multi-layer perceptron structure,it outperforms the best previously existing models in accuracy,while being efficiently trainable on general datasets.We propose a method based on this model for SC forecasting.Using a trained model,we predict the test set from SC 19 to SC 25 with an average mean absolute percentage error of 32.02,root mean square error of 30.3,mean absolute error of 23.32,and R^(2)(coefficient of determination)of 0.76,outperforming other deep learning models in terms of accuracy and training efficiency on sunspot number datasets.Subsequently,we use it to predict the peaks of SC 25 and SC 26.For SC 25,the peak time has ended,but a stronger peak is predicted for SC 26,of 199.3,within a range of 170.8-221.9,projected to occur during April 2034.
基金supported by National Key R&D Program of China(No.2022YFC2404604)Chongqing Research Institution Performance Incentive Guidance Special Project(No.CSTB2023JXJL-YFX0080)Chongqing Medical Scientific Research Project(Joint project of Chongqing Health Commission and Science and Technology Bureau)(No.2022DBXM005)。
文摘This study aimed to integrate Monte Carlo(MC)simulation with deep learning(DL)-based denoising techniques to achieve fast and accurate prediction of high-quality electronic portal imaging device(EPID)transmission dose(TD)for patientspecific quality assurance(PSQA).A total of 100 lung cases were used to obtain the noisy EPID TD by the ARCHER MC code under four kinds of particle numbers(1×10^(6),1×10^(7),1×10^(8)and 1×10^(9)),and the original EPID TD was denoised by the SUNet neural network.The denoised EPID TD was assessed both qualitatively and quantitatively using the structural similarity(SSIM),peak signal-to-noise ratio(PSNR),and gamma passing rate(GPR)with respect to 1×10^(9)as a reference.The computation times for both the MC simulation and DL-based denoising were recorded.As the number of particles increased,both the quality of the noisy EPID TD and computation time increased significantly(1×10^(6):1.12 s,1×10^(7):1.72 s,1×10^(8):8.62 s,and 1×10^(9):73.89 s).In contrast,the DL-based denoising time remained at 0.13-0.16 s.The denoised EPID TD shows a smoother visual appearance and profile curves,but differences between 1×10^(6)and 1×10^(9)still remain.SSIM improves from 0.61 to 0.95 for 1×10^(6),0.70 to 0.96 for 1×10^(7),and 0.90 to 0.97 for 1×10^(8).PSNR increases by>20%for 1×10^(6)and 1×10^(7),and>10%for 1×10^(8).GPR improves from 48.47%to 89.10%for 1×10^(6),61.04%to 94.35%for 1×10^(7),and 91.88%to 99.55%for 1×10^(8).The method that combines MC simulation with DL-based denoising for EPID TD generation can accelerate TD prediction and maintain high accuracy,offering a promising solution for efficient PSQA.
基金supported by the National Key R&D Program of China(No.2023YFA1507500)the Nation-al Natural Science Foundation of China(No.52373159)。
文摘Green solvent pretreatment of biomass represents a promising ap-proach for enhancing the econom-ic value of lignocellulosic deriva-tives.In this study,corncob biomass was treated with a diol-based deep eutectic solvent(DES)under mild conditions,facilitating efficient cellulose separation.The extracted cellulose was subsequently used to fabricate cellulose hydrogels in an aqueous zinc chloride solution.The resulting hydrogel exhibited a“water-in-salt”effect due to the high concentration of ZnCl_(2).Leveraging the antifreeze properties of sorbitol,the system demon-strated outstanding low-temperature electrochemical performance,including a broad operat-ing voltage window and an ionic conductivity of 38.4 mS·cm^(-1)at-20℃.At 20℃,the de-vice achieved an energy density of 206 Wh·kg^(-1)and a power density of 2701.05 W·kg^(-1)at a current density of 1 A·g^(-1).Moreover,the flexible zinc-ion hybrid supercapacitor(ZHSC)maintained 89%of its capacitance and nearly 100%Coulombic efficiency after 5500 cycles at 20℃.This work not only advances the development of zinc-ion energy storage devices but al-so establishes a new paradigm for the green and direct utilization of biomass-derived materi-als.
基金the National Natural Science Foundation of China Youth Fund(NO:322019447)Yulin city 2nd Science and Technology Light“Scientist+Engineer”Team Project(2024-KJZG-K+G-017)Shaanxi Science and Technology Innovation Team Project(2024RS-CXTD-70)for the financial support.
文摘Carnauba wax-structured soybean oil oleogels were developed as alternative frying media for proso millet rice cake processing.Optimal practical properties were identified in the 8%wax concentration through evaluations conducted via polarized light microscopy,oil holding capacity,freeze-thaw stability,and rheological analysis.Oleogel-fried cakes were found to exhibit 6.69%–19.98%lower oil absorption compared to soybean oil-fried samples during 0–18 h frying.The corresponding low-field NMR results demonstrated that the percentage of the relaxation peak area attributed to oil content in oleogel-fried proso millet rice cakes was also reduced.The lipid oxidation rates were significantly reduced in oleogels(P<0.05),with total polar compounds(9.89%–22.81%)and acid values(4.21%–24.37%)decreased while thermal stability was maintained.Saturated fatty acid levels in oleogel were reduced by 3.18%–7.26%after frying for 12–18 h.Texture properties were slightly affected whereas color attributes were significantly improved.Enhanced texture,flavor,hardness,and mouthfeel in oleogel-fried products were confirmed through sensory evaluation.Volatile components,particularly ketones and esters were significantly increased in rice cakes fried with oleogel.These results might offer novel insights for advancement of future low-oil frying technologies and expand the application of oleogels.
