Explore the vast range of titles available.

Male infertility is a global health issue, with 40–50% attributed to sperm abnormalities. The subjectivity and irreproducibility of existing detection methods pose challenges to sperm assessment, making the design of automated semen analysis algorithms crucial for enhancing the reliability of sperm evaluations. This paper proposes a comprehensive sperm tracking algorithm (Sperm YOLOv8E-TrackEVD) that combines an enhanced YOLOv8 small object detection algorithm (SpermYOLOv8-E) with an improved DeepOCSORT tracking algorithm (SpermTrack-EVD) to detect human sperm in a microscopic field of view and track healthy sperm in a sample in a short period effectively. Firstly, we trained the improved YOLOv8 model on the VISEM-Tracking dataset for accurate sperm detection. To enhance the detection of small sperm objects, we introduced an attention mechanism, added a small object detection layer, and integrated the SPDConv and Detect_DyHead modules. Furthermore, we used a new distance metric method and chose IoU loss calculation. Ultimately, we achieved a 1.3% increase in precision, a 1.4% increase in recall rate, and a 2.0% improvement in mAP@0.5:0.95. We applied SpermYOLOv8-E combined with SpermTrack-EVD for sperm tracking. On the VISEM-Tracking dataset, we achieved 74.303% HOTA and 71.167% MOTA. These results show the effectiveness of the designed Sperm YOLOv8E-TrackEVD approach in sperm tracking scenarios.

In large-scale offshore wind power transmission systems, the offshore converter valves are typically based on the half-bridge Modular Multilevel Converter (MMC) topology. This design leads to considerable weight and high costs, presenting a critical bottleneck for the development of offshore wind power transmission. This paper proposes a hybrid topology consisting of paralleled MMCs connected in series with a Diode Rectifier Unit (DRU) to achieve lightweight offshore converter valves. The parallel configuration enhances the steady-state current-carrying capacity of the valve group to match the DRU valve group, and power balance among the paralleled MMCs is realized through an additional DC current-sharing control loop. A calculation method for the main circuit parameters of this lightweight topology is presented, along with a complete parameter calculation process. A design example based on actual engineering capacity is provided. PSCAD simulation results verify that the electrical quantities during steady-state operation of the hybrid topology are consistent with the designed parameters, confirming the correctness of the proposed parameter calculation method.

The problem of the relative position coordinated control for spacecraft formation flying with a leader spacecraft under the obstacle environment is the focus of this paper. To avoid obstacle/collision and maintain the formation configuration, the Null-Space-Based behavioral control architecture is built by defining the priorities of the basic tasks and computing the corresponding velocity vectors. Through the null-space projection, the desired velocity of each follower spacecraft can be calculated by merging the basic tasks. Moreover, due to the partial access to the dynamic leader spacecraft’s states, the distributed estimators are presented for each follower spacecraft. Then, based on the desired velocity, the adaptive coordinated tracking control algorithm incorporated with the barrier Lyapunov function is designed such that the states satisfy the time-varying constraints, even subject to uncertainties and unknown disturbances. Finally, numerical simulations are performed to illustrate the main results.

We report the generation of human ESC-derived, expandable hepatic organoids (hEHOs) using our newly established method with wholly defined (serum-free, feeder free) media. The hEHOs stably maintain phenotypic features of bipotential liver stem/progenitor cells that can differentiate into functional hepatocytes or cholangiocytes. The hEHOs can expand for 20 passages enabling large scale expansion to cell numbers requisite for industry or clinical programs. The cells from hEHOs display remarkable repopulation capacity in injured livers of FRG mice following transplantation, and they differentiate in vivo into mature hepatocytes. If implanted into the epididymal fat pads of immune-deficient mice, they do not generate non-hepatic lineages and have no tendency to form teratomas. We further develop a derivative model by incorporating human fetal liver mesenchymal cells (hFLMCs) into the hEHOs, referred to as hFLMC/hEHO, which can model alcoholic liver disease-associated pathophysiologic changes, including oxidative stress generation, steatosis, inflammatory mediators release and fibrosis, under ethanol treatment. Our work demonstrates that the hEHOs have considerable potential to be a novel, ex vivo pathophysiological model for studying alcoholic liver disease as well as a promising cellular source for treating human liver diseases.

As the restaurant industry is facing labor shortage issues, the use of meal delivery robots instead of waiters/waitresses not only allows the customers to experience the impact of robot technology but also benefits the restaurant business financially by reducing labor costs. Most existing meal delivery robots employ magnetic navigation technologies, which require magnetic strip installation and changes to the restaurant decor. Once the moving path is changed, the magnetic strips need to be re-laid. This study proposes multisource information fusion, i.e., the fusion of ultra-wide band positioning technology with an odometer and a low-cost gyroscope accelerometer, to achieve the positioning of a non-rail meal delivery robot with navigation. By using a low-cost electronic compass and gyroscope accelerometer, the delivery robot can move along a fixed orbit in a flexible and cost-effective manner with steering control. Ultra-wide band (UWB) and track estimation algorithm are combined by extended Kalman filter (EKF), and the positioning error after fusion is about 15 cm, which is accepted by restaurants. In summary, the proposed approach has some potential for commercial applications.

A new virtual synchronous generator (VSG) control strategy was researched and proposed for a VSC-HVDC (High Voltage Direct Current Based on Voltage Source Converter) transmission system. It can be applied to half-bridge or full-half-bridge hybrid topology modular multi-level converter (MMC) to improve the stability and reliability of the system. First, it is proposed that the energy stored in the equivalent capacitor of MMC power module was used to imitate the rotor inertial of synchronous generator. It can buffer transient power fluctuations and synchronize autonomously with the grid. Then the impedance characteristics of the proposed control method have been deduced and analyzed. The results show that the VSG control loop mainly improves the low frequency characteristics of the converter. Secondly, the ability to suppress transient fault current is weak. So, a method, that the given values of inner current loop are calculated by grid impedance matrix, was used. A double closed loop control structure composed by a power outer loop based on VSG control and a current inner loop is obtained. The simulation results show that it can effectively improve the current control capability during the transient process for systems with a 1:2 ratio of converter capacity to grid capacity (The grid short-circuit capacity is 60MW and the MMC is 30 MW). Finally, a hybrid MMC simulation model was built based on PSCAD and the steady-state and transient fault ride-through simulations were performed. The power adjustment time of MMC under the proposed VSG control is about 1s, while the adjustment time under the conventional control strategy is greater than 4s.

A new virtual synchronous generator (VSG) control strategy was researched and proposed for a VSC-HVDC (High Voltage Direct Current Based on Voltage Source Converter) transmission system. It can be applied to half-bridge or full-half-bridge hybrid topology modular multi-level converter (MMC) to improve the stability and reliability of the system. First, it is proposed that the energy stored in the equivalent capacitor of MMC power module was used to imitate the rotor inertial of synchronous generator. It can buffer transient power fluctuations and synchronize autonomously with the grid. Then the impedance characteristics of the proposed control method have been deduced and analyzed. The results show that the VSG control loop mainly improves the low frequency characteristics of the converter. Secondly, the ability to suppress transient fault current is weak. So, a method, that the given values of inner current loop are calculated by grid impedance matrix, was used. A double closed loop control structure composed by a power outer loop based on VSG control and a current inner loop is obtained. The simulation results show that it can effectively improve the current control capability during the transient process for systems with a 1:2 ratio of converter capacity to grid capacity (The grid short-circuit capacity is 60MW and the MMC is 30 MW). Finally, a hybrid MMC simulation model was built based on PSCAD and the steady-state and transient fault ride-through simulations were performed. The power adjustment time of MMC under the proposed VSG control is about 1s, while the adjustment time under the conventional control strategy is greater than 4s.

Flight data anomaly detection is crucial for ensuring the safe operation of unmanned aerial vehicles (UAVs) and has been extensively studied. However, the accurate modeling and analysis of flight data is challenging due to the influence of random noise. Meanwhile, existing methods are often inadequate in parameter selection and feature extraction when dealing with large-scale and high-dimensional flight data. This paper proposes a data-driven multivariate regression-based framework considering spatio-temporal correlation for UAV flight data anomaly detection and recovery, which integrates the techniques of correlation analysis (CA), one-dimensional convolutional neural network and long short-term memory (1D CNN-LSTM), and error filtering (EF), named CA-1DCL-EF. Specifically, CA is first performed on original UAV flight data to select parameters with correlation to reduce the model input and avoid the negative impact of irrelevant parameters on the model. Next, a regression model based on 1D CNN-LSTM is designed to fully extract the spatio-temporal features of UAV flight data and realize parameter mapping. Then, to overcome the effect of random noise, a filtering technique is introduced to smooth the errors to improve the anomaly detection performance. Finally, two common anomaly types are injected into real UAV flight datasets to verify the effectiveness of the proposed method. Graphical Abstract Graphical Abstract

The application of deep neural networks for the semantic segmentation of remote sensing images is a significant research area within the field of the intelligent interpretation of remote sensing data. The semantic segmentation of remote sensing images holds great practical value in urban planning, disaster assessment, the estimation of carbon sinks, and other related fields. With the continuous advancement of remote sensing technology, the spatial resolution of remote sensing images is gradually increasing. This increase in resolution brings about challenges such as significant changes in the scale of ground objects, redundant information, and irregular shapes within remote sensing images. Current methods leverage Transformers to capture global long-range dependencies. However, the use of Transformers introduces higher computational complexity and is prone to losing local details. In this paper, we propose UNeXt (UNet+ConvNeXt+Transformer), a real-time semantic segmentation model tailored for high-resolution remote sensing images. To achieve efficient segmentation, UNeXt uses the lightweight ConvNeXt-T as the encoder and a lightweight decoder, Transnext, which combines a Transformer and CNN (Convolutional Neural Networks) to capture global information while avoiding the loss of local details. Furthermore, in order to more effectively utilize spatial and channel information, we propose a SCFB (SC Feature Fuse Block) to reduce computational complexity while enhancing the model’s recognition of complex scenes. A series of ablation experiments and comprehensive comparative experiments demonstrate that our method not only runs faster than state-of-the-art (SOTA) lightweight models but also achieves higher accuracy. Specifically, our proposed UNeXt achieves 85.2% and 82.9% mIoUs on the Vaihingen and Gaofen5 (GID5) datasets, respectively, while maintaining 97 fps for 512 × 512 inputs on a single NVIDIA GTX 4090 GPU, outperforming other SOTA methods.

For a SLAM system operating in a dynamic indoor environment, its position estimation accuracy and visual odometer stability could be reduced because the system can be easily affected by moving obstacles. In this paper, a visual SLAM algorithm based on the Yolov4-Tiny network is proposed. Meanwhile, a dynamic feature point elimination strategy based on the traditional ORBSLAM is proposed. Besides this, to obtain semantic information, object detection is carried out when the feature points of the image are extracted. In addition, the epipolar geometry algorithm and the LK optical flow method are employed to detect dynamic objects. The dynamic feature points are removed in the tracking thread, and only the static feature points are used to estimate the position of the camera. The proposed method is evaluated on the TUM dataset. The experimental results show that, compared with ORB-SLAM2, our algorithm improves the camera position estimation accuracy by 93.35% in a highly dynamic environment. Additionally, the average time needed by our algorithm to process an image frame in the tracking thread is 21.49 ms, achieving real-time performance.