Lithium-ion batteries have become a dominant energy storage technology, widely used in critical applications such as electric vehicles, consumer electronics, and large-scale energy storage systems. As a key component, electrode sheets significantly influence battery performance and safety; defects occurring during their manufacturing process can degrade functionality and pose potential safety risks. Therefore, accurate and efficient defect detection during electrode sheet production is essential. However, the presence of multiple defect types and the high similarity among defect features often cause conventional detection networks to suffer from missed or false detections. To overcome these challenges, this paper proposes an enhanced detection network, termed YOLOv8s_AD. First, DSCDownsample utilizes convolutional layers with varying receptive fields to capture multi-scale features, thereby minimizing information loss during downsampling and preserving critical feature details. Second, the C2f_CDW module is deployed in shallow layers to extract local spatial features, while the C2f_CIB module is applied in deeper layers to capture global contextual information. This hierarchical design reduces feature redundancy and enhances the capability to distinguish diverse defect patterns. Finally, the proposed Cross Neck structure in the neck facilitates comprehensive cross-scale and cross-location feature interactions, enabling global fusion of information from multiple network layers to improve representation capacity. Experimental results demonstrate that YOLOv8s_AD reduces the number of parameters and GFLOPs by 43.5% and 41.4% respectively, while increasing mAP@50 and mAP@[50:-95] by 0.4% and 4.5%. These improvements indicate that the proposed network achieves superior detection performance with reduced computational complexity.

Fully-connected tensor network (FCTN) has recently drawn lots of attention in tensor completion due to its full description of all correlations between any two modes. However, the FCTN model has multiple ranks, and existing methods often ignore the difficulty brought about by rank selection, especially when the model is rank-sensitive. To overcome this drawback, this paper proposed a low-rank sparse FCTN for tensor completion. Specifically, we theoretically show that, for a tensor with FCTN structure, its subtensor can be represented as a coefficient sum of basic tensors, where the coefficients are remained in the FCTN’s factors. This means that factors’ sparsity can improve the robustness of FCTN to larger rank selection. Moreover, according to the relation between the target tensor and its FCTN’s factors, low-rank constrain is used to enhance rank robustness. Lastly, we optimize the proposed model by the alternating direction method of multipliers (ADMM) algorithm. Experimental results show that the low-rank sparse constraint effectively improves the rank robustness of the FCTN model, and achieves excellent results compared with other state-of-the-art completion methods.

Tensor wheel decomposition has recently drawn lots of attentions in tensor completion, due to its advantages of wheel topology in exploring the intrinsic relationships. However, since the rank of tensor wheel is defined as a vector, it is very hard to select one rather-good rank for tensor completion when the model is rank-sensitive, i.e., the model is prone to overfitting due to rank selection. To solve this problem, under the tensor wheel structure, we theoretically analyze the relationship of sparsity and smoothness to the overfitting, which is expected to improve the performance by preventing the overfitting due to excessive rank selection. Then, based on the analysis of sparsity and smoothness, we proposed a novel tensor wheel completion model with sparsity and smoothness on latent space. Lastly, an efficient alternating direction method of multipliers (ADMM)-based algorithm is developed to optimize the proposed model. Experimental results show that the proposed method is superior to some existing methods in tensor completion and can maintain good results in a large range of rank selection, which enable the proposed method is not easy to overfit with the increasing of rank.

With the rapid advancements in automation and soft robotics, the exploration of mobile robots for applications in complex environments is increasingly deepening. This paper presents a novel dual soft arm mobile robot (DSAMR), whose design integrates advanced soft robotics technologies with biomimetic design inspired by human arms, aiming to achieve efficient obstacle avoidance and object manipulation. The robot employs Bubble Artificial Muscle Arms (BAMAs) for locomotion, enabling flexible movements such as forward, backward, and turning motions; it also integrates TacTip (tactile fingertip), a biomimetic sensor that mimics the tactile structure of human fingertips, to achieve real-time perception. BAMAs and TacTip collaborate to achieve the integration of perception and operation like a human hand, enabling the system to accurately detect obstacles and manipulate objects, including typical delicate items such as a paper towel roll and a pen, with the maximum capacity to grasp objects weighing up to 148.8 g. Experiments have demonstrated that a single inflation-deflation cycle of the BAMAs enables the DSAMR to turn right by 35.5° and left by 28.3°, and successfully allows the DSAMR to recognize obstacles and turn to avoid them. The experimental results indicate that the DSAMR can operate effectively in dynamic environments, with excellent stability and obstacle avoidance capabilities. This paper discusses the design details of BAMA actuators, steering engines, and TacTip, as well as their integration into the robot's motion and sensing systems. The findings emphasize the DSAMR's potential applications in industrial automation, particularly in the context of Industry 4.0. Finally, the study summarizes optimization strategies and future improvement directions to enhance the robot's operational efficiency, including onboard power integration and advanced obstacle recognition technologies.

Surface defects on lithium battery electrodes are typically small and diverse, which will impact battery performance and safety. Due to their morphological characteristics, traditional detection methods often struggle to accurately identify tiny and highly similar defects. Additionally, the existing YOLOv8 model faces some challenges in defect detection on lithium battery electrodes, such as large model parameters, insufficient feature extraction, and inadequate feature fusion. To address these issues, this paper proposes a lightweight detection model, H2LF-YOLOv8n. First, the GSS and MSEF modules are introduced in the backbone network. The GSS module enhances small defect feature extraction by regulating and interacting information between channels, while also reducing the number of parameters. The MSEF module uses a multi-branch structure to extract and fuse channel information at different scales, thereby improving the capture of multi-scale features. Second, the C2f_Faster structure is proposed in the feature fusion stage. It employs partial convolution for sparse computation and a channel fusion strategy, while integrating the PSA mechanism to dynamically select defect-sensitive regions. This combination facilitates efficient information interaction and feature fusion. Finally, due to the fact that the surface defects of lithium battery electrodes are extremely small in size, redundant detection heads are removed to reduce the model's parameter size. Experimental results demonstrate that the improved H2LF-YOLOv8n maintains high detection accuracy while reducing the parameter size by 38.7% compared to the original YOLOv8n. This makes the model more suitable for practical defect detection on lithium battery electrodes in industrial production environments.

Image restoration seeks to improve degraded images, recovering high-quality versions. However, existing model-based methods are limited by predefined priors, and suffered with suboptimal denoising performance, while deep plug-and-play denoisers often suffer from detail loss and weak noise generalization. To overcome these limitations, we propose SRUNet++, a plug-and-play image restoration network. Firstly, multi-scale features are aggregated via dense skip connections to integrate low-frequency semantics and high-frequency details. Secondly, a multi-layer residual series (MLRS) module is introduced to enhance local feature reuse and suppress noise interference. Finally, training with observed images and large-scale random noise level maps as input boosts the denoising and generalization of the model. Experimental results demonstrate that SRUNet++ significantly outperforms existing methods in both image denoising and deblurring tasks.

Due to the complexity of multi-objective engineering problems, solutions obtained by many algorithms often exhibit poor distribution, and the algorithms tend to fall into local optima. To effectively alleviate these issues, an improved multi-objective moth–flame optimization algorithm (IETMFO) is proposed in this paper, with three core novelties: A hybrid mutation mechanism (integrating two mutation techniques) is first used to generate a new population, and then an indicator-based selection mechanism is adopted to produce a high-quality initial population, enhancing solution distribution. Enhanced Brownian motion is introduced as a local search strategy to reduce the risk of falling into local optima. An improved flame update mechanism is incorporated to maintain flame diversity, boosting the algorithm’s adaptability. The IETMFO is tested on 49 benchmark functions and 6 constrained engineering problems, and then compared with seven well-known algorithms (including NSGA-II, MOEA/D, and traditional MFO). The experimental results show the following: in benchmark function tests, IETMFO reduces the IGD value by an average of 32.7% and increases the HV value by an average of 28.5% compared with NSGA-II; on ZDT series functions, it outperforms the seven contrast algorithms in solution distribution uniformity; in the six engineering problems, its optimal solution proportion reaches 66.7%, and the risk of falling into local optima is reduced by 41.2%. These results demonstrate that the IETMFO achieves competitive performance in addressing multi-objective engineering problems.

This paper proposes an adaptive constrained control considering input saturation and prescribed performance constraints for uncertain two-degree-of-freedom helicopter systems with actuator faults. A devised performance function precisely indicates the time for the system to achieve stability and adjusts the convergence speed of its dynamic process through parameter modifications. The smoothing Gaussian error function is introduced to replace the saturated nonlinearity of each non-smooth actuator, the radial basis function neural network is employed for approximating the unknown nonlinear function, adaptive auxiliary parameters are introduced, and an adaptive controller is designed to address issues such as model uncertainty, input saturation, and actuator faults arising in practical applications. Furthermore, the Lyapunov direct method is employed to demonstrate the stability and boundedness of the helicopter system. Results obtained from simulations and experiments on the Quanser two-degree-of-freedom helicopter experimental platform confirm the effectiveness and feasibility of the proposed control method.

Traffic sign detection and recognition facilitates real-time monitoring and interpretation of various traffic signs on the road, such as those indicating speed limits, prohibition of overtaking, and navigation cues. This has substantial applications for autonomous driving and decision-making systems. Consequently, designing accurate and efficient algorithms for the automatic recognition of traffic signs is crucial in the intelligent transportation field. However, targets that need to be detected by traffic sign recognition applications are mostly small-sized, causing challenges regarding their automatic recognition. The YOLOv5s model, characterized by its minimal depth and narrowest feature map, has gained widespread popularity for executing detection owing to its features of being lightweight and easily portable. Furthermore, the YOLOv5s model uses an anchor-based prediction approach that uses anchor boxes of different sizes and shapes to regress and classify various targets. This method generates dense anchor boxes and enables the model to directly perform object classification and bounding box regression, thereby enhancing its target recall capability. Therefore, the anchor-based Yolov5s method has been applied to traffic sign detection; however, it suffers from issues such as false positives and missed detection. Detection of small targets continues to be a challenging aspect in current traffic sign recognition technology due to the following: small targets carry less information; detection of small targets requires high precision in positioning; and environmental noise may overwhelm the detection of small targets. To overcome the abovementioned issues, such as missed detection, false positives, and low detection accuracy, this study proposes a model called STD-YOLOv5s that is specifically designed for small target detection. First, by increasing the number of upsampling and prediction output layers, this model obtains abundant location information. This can enhance the global understanding of images and solve the issue of insufficient information associated with small targets. Second, the CA attention mechanism is added after each C3 module, whereas the Swin-T attention mechanism module is added before each output layer, increasing the model's ability to capture multilayer feature information and consequently improving its performance of small target detection. Finally, the accuracy of target localization is ensured using the SIoU penalty function, which considers the target shape and spatial relationships, thereby increasing the model's ability to capture the positional relationships among targets of different sizes in the image. The STD-YOLOv5s model was validated using the TT100K dataset by ablation and comparison experiments. Experimental results indicate that the proposed model not only maintains the lightweight nature and high detection speed of the YOLOv5s model but also achieves improvements in precision, recall, and average precision.

The Central Asian Orogenic Belt underwent complex tectonic processes and is one of the most intensely accretionary areas globally. Porphyry copper deposits within the belt were likely subjected to deformation during tectonic processes. The Bainaimiao Cu-Au deposit is a typical example of a deformed porphyry deposit whose formation processes include a porphyry emplacement (Event I), a greenschist facies metamorphism (Event II), and a brittle deformation (Event III). Geochronology and trace element geochemistry of zircon, volatiles of magmatic apatite, along with the assemblages, textures, abundances, and compositions of phyllosilicates from these three events were investigated to unveil the physicochemical conditions under which the key geological events relevant to the deposit formation took place. LA-ICP-MS zircon U-Pb dating shows that the granodiorite porphyry in the northern and southern zones formed at 447.4 +/- 1.6 to 445.8 +/- 3.6 Ma and 436.1 +/- 3.8 to 434.1 +/- 3.3 Ma, respectively. The granodiorite porphyry in the northern zone has higher oxygen fugacity and Clmelt content but similar Smelt and Fmelt contents compared to the granodiorite porphyry in the southern zone. Microscopic and mineralogic observations point to Event I to be of high plagioclase (22-67 vol%) and quartz (6-40 vol%) content with a range of hydrothermal minerals related to potassic, phyllic, and propylitic alterations. Event II features high amphibole (38-83 vol%) or epidote-chlorite (up to 77 vol%) content with minerals precipitating along the schistosity planes. Event III is characterized by wide veins (3-80 cm) and the highest quartz (61-65 vol%) and calcite (12-19 vol%) content. Geothermometry results show the temperature of potassic and phyllic alterations of Event Ito be 622 degrees C and 288 degrees C, respectively. Based on geothermometry and P-T pseudosections, the temperatures of metamorphism and metallic precipitation of Event II were 271-634 degrees C and 297-328 degrees C, respectively. Both mechanical and chemical mobilization of metallic elements results in Cu mineralization during Event II. The metallic precipitation temperatures of Event III spanned from 297 to 328 degrees C according to chlorite geothermometry. The ratios of Fe3+/Fetotal and Mg/(Mg + Fetotal) of biotite, chlorite Fe/(Fe + Mg), and white K-mica composition show the mineralizing fluid of Event III to be the most oxidized while that of Event II is the most reduced, F-rich and features the lowest water/rock ratio. This study suggests that deformation processes can increase the Cu mineralization grade of the deformed porphyry deposits through mobilization and re-precipitation of metallic elements.

多通道欠定卷积语音混合信号的分离问题是盲信号分离领域的难点.由于混合信号中常伴随声学回声和混响，导致真实的源信号很难完全被清晰地分离出来，传统的盲信号分离算法多数适用于低混响，而在高混响场景下，算法的分离性能极速下降甚至是失效的.本文针对具有声学回声和混响环境下的多通道欠定卷积语音混合信号的分离问题，提出一种结合脉冲响应重塑和期望最大化的盲信号分离算法，该算法在低混响和高混响下都表现出很好的分离性能.首先，利用基于无穷范数和p-范数的脉冲响应重塑技术设计预滤波器消除可听回声，完成对混合信号的重塑，提高混合信号的质量.然后，对重塑后的混合信号利用分层聚类方法估计混合矩阵，基于期望最大化算法框架，设计新的模型参数实时更新规则，通过结合脉冲响应重塑和期望最大化重构源信号.实验结果表明，所提算法可以有效地分离不同混响环境下带声学回声的欠定卷积混合信号，其分离性能优越，同时对噪声具有很好的鲁棒性.

In this paper, an active fault-tolerance controller is proposed for a two-degree-of-freedom (2-DOF) helicopter model with different types of sensor faults. The proposed controller is designed by combining an interval type-2 fuzzy logic control approach with a sliding-mode control technique. In addition, the controller is compensated by a sliding-mode observer used to estimate faults in real-time. For the proposed control scheme, the overall stability of the closed-loop system is proven using the Lyapunov stability theory. Finally, numerical simulations and experiments verify that the proposed scheme achieves superior control performance in the case of sensor failure.

This study develops a neural network control for a non-linear two-degree-of-freedom unmanned helicopter system satisfying prescribed performance constraints is developed. By introducing a prescribed performance function to express the system constraints, a constrained tracking control problem is transformed into an equivalent unconstrained stability problem through error transformations. The neural network controller is designed to ensure that the tracking error converges to a small neighborhood of zero and that the prescribed performance constraints are satisfied. Furthermore, the stability and error convergence of the system are analysed through the Lyapunov direct approach. The effectiveness of the proposed control scheme is verified by simulation and experimental results.

In this paper, an adaptive model-free optimal reinforcement learning (RL) neural network (NN) control scheme based on filter error is proposed for the trajectory tracking control problem of an autonomous underwater vehicle (AUV) with input saturation. Generally, the optimal control is realized by solving the Hamilton-Jacobi-Bellman (HJB) equation. However, due to its inherent nonlinearity and complexity, the HJB equation of AUV dynamics is challenging to solve. To deal with this problem, an RL strategy based on an actor-critic framework is proposed to approximate the solution of the HJB equation, where actor and critic NNs are used to perform control behavior and evaluate control performance, respectively. In addition, for the AUV system with the second-order strict-feedback dynamic model, the optimal controller design method based on filtering errors is proposed for the first time to simplify the controller design and accelerate the response speed of the system. Then, to solve the model-dependent problem, an extended state observer (ESO) is designed to estimate the unknown nonlinear dynamics, and an adaptive law is designed to estimate the unknown model parameters. To deal with the input saturation, an auxiliary variable system is utilized in the control law. The strict Lyapunov analysis guarantees that all signals of the system are semi-global uniformly ultimately bounded (SGUUB). Finally, the superiority of the proposed method is verified by comparative experiments.

In this paper, a reinforcement learning (RL) fault-tolerant control (FTC) method is proposed for trajectory tracking of autonomous underwater vehicles (AUVs) with thruster faults. To deal with the thruster fault, unknown disturbance and model uncertainty, a new integral extended state observer (IESO) for fault diagnosis observation is proposed, which uses a conventional ESO to estimate the total system uncertainty, and introduces an integral mechanism to mitigate the effect of estimation error further. Thus, the problem that the estimation error caused by the traditional ESO leads to the decline of the fault-tolerant capability of the FTC system is solved. Then, to solve the problem of integral saturation due to the introduction of the integral term, the integral term is limited after the thruster fault of the AUV. Furthermore, based on the actor–critic structure of RL, a PD-like feedback controller is designed to realize the FTC of AUV in the face of thruster fault by using the total uncertainty of the IESO scheme. And the input saturation of the thruster is considered, and an auxiliary variable system is used to handle the control truncation between saturated and unsaturated inputs. Based on the Lyapunov method, the stability of the closed-loop system is analyzed and proved. Finally, the proposed method is verified to have good fault tolerance and robustness by simulation and underwater experiments.

In this article, we propose a new adaptive fuzzy fault-tolerant control (FTC) for a three-dimensional riser-vessel system with unknown backlash nonlinearity. A model for the smooth inverse dynamics of the backlash is introduced; then, the control input is divided into an expected input and a compensation error. Considering the imprecision of system modeling and unknown external disturbances, we employ a fuzzy adaptive technology to achieve compensation. By incorporating the actuator fault term and backlash error, the adaptive FTC is developed to resolve loss faults in the actuator and compensate for the unknown backlash to some extent. The direct Lyapunov method is used to demonstrate the system’s bounded stability. Finally, simulation results demonstrate the effectiveness of the derived scheme.

Aiming at the low processing speed of the Sobel edge detection algorithm and the equipment limitations of Compute Unified Device Architecture (CUDA) implementation algorithm acceleration, a Sobel edge detection parallel algorithm based on Open Computing Language (OpenCL) architecture is proposed. The algorithm uses the heterogeneous mode of CPU + GPU to achieve algorithm acceleration. According to the parallel structure and hardware characteristics of Graphics Processing Unit (GPU), the parallel algorithm adopts two acceleration technologies, multi-level storage technology, and vector access technology, which optimizes the data storage structure, improves the data access efficiency, and reduces the complexity of the algorithm. Unlike the CUDA implementation algorithm acceleration for NVIDIA graphics card devices, the OpenCL parallel improved algorithm has no device limitations. Experimental results show that compared with the CPU serial algorithm, the OpenMP parallel algorithm, and the CUDA parallel algorithm, the parallel algorithm has obtained 9.55 times, 2.23 times, and 1.17 times speedup, respectively. The parallel algorithm in this paper shows good data expansibility and platform portability and can provide technical support for the deep application of massive image data.

Low-cost MnO2 has been vigorously investigated in the field of catalysis. However, high amount of MnO2 usually needs to be used in the disinfection process owing to their low photocatalytic activity. In this work, modification of MnO2 with zeolitic imidazole framework-8 (ZIF-8) for improving the photocatalytic bactericidal performance was explored for the first time. Specifically, MnO2/ZIF-8 exhibited complete inactivation of multi-drug resistant Escherichia coli (E. coli) at a low concentration (3.24 mu g/mL) with simulated solar irradiation. The 1O2 rather than center dot O2xe213; or center dot OH was identified as the major reactive species for the bacterial inactivation. Mechanistic studies indicated that the superior bactericidal property of MnO2/ZIF-8 than the pristine material could be ascribed to the increased surface oxygen vacancies (OVs) amount and an inhibited electron and hole recombination. This study provides a novel type of photocatalyst for antibacterial application and may shed light on designing MnO2 based low-cost nanocomposites with photocatalytic bactericidal capability for the practical water disinfection treatment.

Image edge is the most indicative feature that forms a significant role in image analysis and image understanding, but edge-detail preservation is a difficult task in image restoration due to noise and blur during imaging. The balance between edge preservation and noise removal has always been a difficult problem in image restoration. This paper proposes a derivative fidelity-based total generalized variation method (D-TGV) to improve this balance. First, an objective function model that highlights the ability to maintain details is proposed for the image restoration problem, which is combined with a fidelity term in derivative space and a total generalized variation regularization term. This is designed to achieve the advantage of preserving details in derivative space and eliminate the staircase effect caused by traditional total variation. Second, the alternating direction method of the multipliers (ADMM) is used to solve the model equations by decomposing the original, highly complex model into several simple sub-problems to attain rapid convergence. Finally, a series of experiments conducted on standard grayscale images showed that the proposed method exhibited a good balance between detail preservation and denoising but also reached completion with the fewest iterations compared with the currently established methods.

The Wulong gold deposit (> 80 t Au) is located at the northeastern margin of the North China Craton (NCC). Gold in the most economically important quartz veins (Stages 2 and 3) is associated with a varied assemblage of Bi and Te minerals. Stage 2 is characterized by, in a temporal order, native gold-native bismuth-maldonite-hedleyite, bismuthinite-Bi-sulfotellurides, electrum-hessite-Bi-Pb-Ag sulfosalts, Bi-Pb sulfosalts, and Bi-Pb-sulfotellurides. All of these minerals except maldonite, electrum, hessite, and Bi-sulfosalts are present in Stage 3 veins, which also contain jonassonite and unnamed minerals with stoichiometry of Bi8Te3 and Bi5Te3. These complex Bi mineral assemblages resulted from changes in Pb, Ag, Te and Bi concentrations in reduced hydrothermal fluids both locally and temporally, from Stage 2 to 3. Texturally, approximately 85% of the gold grains are associated with almost all Bi minerals in each main-stage mineralization. Quartz-vein ores from Stages 2 and 3 have high Bi concentrations (up to 7332 ppm) and show a positive correlation between Bi and Au, with Bi/Au ratios > 10. These textural and geochemical associations, together with the occurrence of numerous droplet-like blebs and small inclusion trails of native bismuth and Bi-tellurides with native gold, maldonite, or jonassonite, suggest that Bi-rich melts may have formed from the hydrothermal fluids of Stages 2 and 3, and acted as an Au scavenger in a complex Bi-Au-Pb-Ag-Te system. Hence, the intimate association of Bi and Au at Wulong is the result of Au enrichment via the liquid bismuth collector mechanism. Unlike Wulong, however, other deposits in the NCC, such as Yangzhaiyu (Xiaoqinling district), Guilaizhuang (Jiaodong district) and Dongping (Jidong district), are characterized by low Bi concentrations, and are much richer in Te than Bi, so these systems could not generate low-melting point Bi-rich polymetallic melts. The low-sulfur Bi-Au-Pb-Ag-Te association at Wulong is consistent with that observed in reduced intrusion-related gold systems elsewhere and suggests that hydrothermal fluids may have been derived from the co-located Early Cretaceous diorite-granite porphyry dyke swarm.