In practical industrial application, the fault samples collected from rotating machinery are frequently unbalanced, which will create difficulties when it comes to diagnosis. Besides, the variation of working conditions and noise factors will further reduce the diagnosis’s accuracy and stability. Considering the above problems, we established a model based on deep Wasserstein generative adversarial network with gradient penalty (DWGANGP). In this model, the unbalanced fault data set will first be trained by the sample generation network to generate synthetic samples, which will be used to restore the balance. A one-dimensional convolutional neural network with a specific structure is then used as the fault diagnosis network to classify the reconstructed equilibrium samples. The experimental results show that the proposed sample generation network can generate high-quality synthetic samples under highly imbalanced data, and the diagnostic network has a fast training convergence. Compared to the combination methods of support vector machines, back propagation neural network and deep belief network, our method has a 74% average accuracy in all unbalanced experimental conditions, which has 64%, 69% and 87% averages leading, respectively.

The paradigm of “deep-learning visual perception + hand–eye transformation + motion planning” for robot grasping has demonstrated viable capabilities in specific scenarios. However, its further development faces challenges in handling complex and dynamic environments. This article proposes a keypoint detection network-driven visual servoing grasping framework. First, we develop an efficient two-stage keypoint detector to perform real-time inference of sparse image-plane features for the target. Subsequently, a low-pass filtering algorithm is employed to smoothen the detected keypoints. These processed keypoints are then used in an image-based visual servoing (IBVS) controller to calculate the robot joint velocities, enabling precise tracking. A specialized dataset for training and evaluation was constructed using domain randomization techniques, comprising 11 K samples across six categories. Comprehensive experiments demonstrate the detector’s low latency and accurate performance, even in low lighting, overexposure, partial occlusion, and densely packed environments. Static and dynamic grasping experiments validate that this framework achieves localization accuracy superior to five pixels and an overall grasping success rate exceeding 70% under unknown hand–eye calibration. The dataset is provided at github.com/hijunqi/VS_grasping_keypoint_detection_dataset.

Predicting the time series of 10.7-cm solar radio flux is a challenging task because of its daily variability. This paper proposed a non-linear method, a convolutional and recurrent neural network combined model to achieve end-to-end F10.7 forecasts. The network consists of a one-dimensional convolutional neural network and a long short-term memory network. The CNN network extracted features from F10.7 original data, then trained the feature signals in the long short-term memory network, and outputted the predicted values. The F10.7 daily data during 2003–2014 are used for the testing set. The mean absolute percentage error values of approximately 2.04%, 2.78%, and 4.66% for 1-day, 3-day, and 7-day forecasts, respectively. The statistical results of evaluating the root mean square error, spearman correlation coefficient shows a superior effect as a whole for the 1–27 days forecast, compared with the ordinary single neural network and combination models.

The daily 10.7-cm Solar Radio Flux (F10.7) data is a time series with highly volatile. The accurate prediction of F10.7 has a great significance in the fields of aerospace and meteorology. At present, the prediction of F10.7 is mainly carried out by linear models, nonlinear models, or a combination of the two. The combination model is a promising strategy, which attempts to benefit from the strength of both. This paper proposes an Empirical Mode Decomposition (EMD) -Long Short-Term Memory Neural Network (LSTMNN) hybrid method, which is assembled by a particular frame, namely EMD–LSTM. The original dataset of F10.7 is firstly processed by EMD and decomposed into a series of components with different frequency characteristics. Then the output values of EMD are respectively fed to a developed LSTM model to acquire the predicted values of each component. The final forecasting values are obtained after a procedure of information reconstruction. The evaluation is undertaken by some statistical evaluation indexes in the cases of 1-27 days ahead and different years. Experimental results show that the proposed method gives superior accuracy as compared with benchmark models, including other isolated algorithms and hybrid methods.

Existing 6-degree-of-freedom (6-DoF) robotic grasping methods based on 3D pose estimation often suffer from long-standing issues of quantization errors, inference delays, and susceptibility to interference. We explore a novel approach that reformulates 6-DoF grasping as a minimizing problem of the projection error of 2D keypoints. We first input the RGB stream from the RGB-D sensor into an efficient single-stage keypoint detector to extract sparse keypoints of the target. Subsequently, we augment the training data using domain randomization to reduce real-world annotation costs and improve robustness to occlusions and extreme lighting. We then employ an Image-Based Visual Servoing (IBVS) controller, integrating keypoint features with the depth data of the RGB-D sensor, to transform keypoint deviations into corresponding joint velocity commands for accurate robot tracking. Real-world grasping experiments demonstrate that our method achieves over 70% grasping success in scenarios with unknown hand-eye calibration. Moreover, it maintains pixel-level accuracy under cluttered static and dynamic conditions.