Subspace clustering plays an important role in the tasks such as data processing and pattern recognition. Since the high-dimensional data may contain complex noise, as well as non-linear structure, learning low-dimensional subspace structures is a challenging task. However, the existing methods to deal with both problems relax the original problem convexly. The results of solving by these methods deviate from the solution of the original problem. In this paper, to overcome this deficiency, we propose a robust low-rank kernel subspace clustering model, which coalesces the non-convex Schatten p-norm $(0 \lt p \le 1)$ regularizer with kernel trick and correntropy. Our kernel trick extends linear subspace clustering to non-linear counterparts, the Schatten p-norm regularizer can approximate the rank of the data in feature space effectively, and the correntropy is a robust measure to large corruptions. Furthermore, an efficient iterative algorithm (HQ-ADMM) is designed to solve the formulated problem, which coalesces the half-quadratic technique and Alternating Direction Method of Multipliers. This algorithm can ensure the closed form solutions at each iteration, which improves the computation speed of the algorithm. Extensive experiments on face/object clustering and motion segmentation clearly attest the ascendancy of the proposed method over several state-of-the-art methods.

The prediction about short-term traffic flow is a crucial research field in Intelligent Transportation Systems (ITS), which focuses on forecasting traffic data to provide data for optimal traffic scheduling to reduce congestion. This paper presents a new deep learning network based on the graph convolution network (GCN) and long short-term memory (LSTM) unit. The model was trained and tested by using local traffic data from the city of Mianyang. After introducing the new deep learning network, a data reduction method is also established in this paper which was select the relevant road links as model’s input based on time correlation. Finally, the proposed algorithm is compared with other popular methods, and better results can be obtained in the conventional 5-minute prediction test. At the same time, the performance of this GCN-LSTM model can be maintained in different time range from 5 minutes to 125 minutes by multi-step prediction, which is much better than other models.

The remaining useful life (RUL) prediction of rolling bearings based on vibration signals has attracted widespread attention. It is not satisfactory to adopt information theory (such as information entropy) to realize RUL prediction for complex vibration signals. Recent research has used more deep learning methods based on the automatic extraction of feature information to replace traditional methods (such as information theory or signal processing) to obtain higher prediction accuracy. Convolutional neural networks (CNNs) based on multi-scale information extraction have demonstrated promising effectiveness. However, the existing multi-scale methods significantly increase the number of model parameters and lack efficient learning mechanisms to distinguish the importance of different scale information. To deal with the issue, the authors of this paper developed a novel feature reuse multi-scale attention residual network (FRMARNet) for the RUL prediction of rolling bearings. Firstly, a cross-channel maximum pooling layer was designed to automatically select the more important information. Secondly, a lightweight feature reuse multi-scale attention unit was developed to extract the multi-scale degradation information in the vibration signals and recalibrate the multi-scale information. Then, end-to-end mapping between the vibration signal and the RUL was established. Finally, extensive experiments were used to demonstrate that the proposed FRMARNet model can improve prediction accuracy while reducing the number of model parameters, and it outperformed other state-of-the-art methods.

Underwater applications present unique challenges such as color deviation, noise, and low contrast, which can degrade image quality. Addressing these issues, we propose a novel approach called the learnable full-frequency transformer dual generative adversarial network (LFT-DGAN). Our method comprises several key innovations. Firstly, we introduce a reversible convolution-based image decomposition technique. This method effectively separates underwater image information into low-, medium-, and high-frequency domains, enabling more thorough feature extraction. Secondly, we employ image channels and spatial similarity to construct a learnable full-frequency domain transformer. This transformer facilitates interaction between different branches of information, enhancing the overall image processing capabilities. Finally, we develop a robust dual-domain discriminator capable of learning spatial and frequency domain characteristics of underwater images. Extensive experimentation demonstrates the superiority of the LFT-DGAN method over state-of-the-art techniques across multiple underwater datasets. Our approach achieves significantly improved quality and evaluation metrics, showcasing its effectiveness in addressing the challenges posed by underwater imaging. The code can be found at https://github.com/zhengshijian1993/LFT-DGAN.

In response to the lack of a unified cyber–physical system framework, which combined the Internet of Things, industrial big data, and deep learning algorithms for the condition monitoring of critical transmission components in a smart production line. In this study, based on the conceptualization of the layers, a novel five-layer cyber–physical systems framework for smart production lines is proposed. This architecture integrates physics and is data-driven. The smart connection layer collects and transmits data, the physical equation modeling layer converts low-value raw data into high-value feature information via signal processing, the machine learning modeling layer realizes condition prediction through a deep learning algorithm, and scientific decision-making and predictive maintenance are completed through a cognition layer and a configuration layer. Case studies on three critical transmission components—spindles, bearings, and gears—are carried out to validate the effectiveness of the proposed framework and hybrid model for condition monitoring. The prediction results of the three datasets show that the system is successful in distinguishing condition, while the short time Fourier transform signal processing and deep residual network deep learning algorithm is superior to that of other models. The proposed framework and approach are scalable and generalizable and lay the foundation for the extension of the model.

An adaptive back-stepping sliding mode controller (ABSMC) algorithm was developed for nonlinear uncertain systems based on a nonlinear disturbance observer (NDO). The developed ABSMC was applied to attitude control for the dual arm of a humanoid robot. Considering the system uncertainty and the unknown external disturbances, the ABSMC scheme was designed to eliminate the chattering phenomenon in the traditional sliding mode control and to reduce the tracking error closer to zero. The ABSMC algorithm solved problems related to the chattering of the system for both uncertainties and disturbances in the humanoid robotic system with an NDO in a two-dimensional environment. The algorithm was designed to work equally well with agents, with higher degrees of freedom in different applications. The method was appropriate for improving tracking performance. The ABSMC algorithm guaranteed global stability and improved the dynamic performance of the system. The algorithm inherited a low computational cost, probabilistic completeness, and asymptotic optimality from the fuzzy sliding mode control. This algorithm has a practical application in the dual arm of a humanoid robot with a circular trajectory. This paper showed the effectiveness and applicability of the proposed methods, which reduced the output of the controller and improved the control performance of the humanoid robotic system. The new combined control algorithm, ABSMC, was able to feasibly and efficiently weaken the chattering on the robot's closed-loop paths, starting and finishing at the same configuration.