Deep-learning-based defect detection models require large and diverse datasets, yet obtaining sufficient defect samples in industrial environments is challenging. Existing sample-generation-based methods alleviate data scarcity, but still depend on a certain amount of training data and struggle to enforce the strict structural specifications of industrial products, limiting their applicability. To address these issues, we propose a sample-generation-based defect detection model. The core of our approach is a defect generation network that employs multiscale progressive learning to extract multiresolution features and enables augmentation from a single sample. To satisfy structural constraints without suppressing diversity, we introduce a structural attention mechanism that guides the generative process rather than imposing direct mask-based restrictions. Additionally, we design a simple yet effective foreground—background reconstruction loss that better preserves structural details compared with conventional reconstruction losses. Our method requires only a single sample to initiate data augmentation. As a result, collecting a small number of representative defect samples is sufficient to significantly enhance detection performance, and the low data requirement allows broad applicability across diverse industrial defect detection tasks. Experimental results demonstrate that our method outperforms existing models, with improved sample quality and detection performance, achieving up to a 28% and 20% increase in [email protected] on the DeepPCB and NEU-DET datasets, respectively. Our method proves even more effective when the sample size is limited.

This study addresses the challenges of hybrid production lines, reconfigurable characteristics, frequent disturbances, and multi-resource constraints in complex aerospace product assembly and testing workshops. We propose a Dynamic Integrated Process Planning and Scheduling under Multi-Resource Constraints in Workshops with Reconfigurable Manufacturing Cells (MRC-DIPPS-RMC). By establishing an integrated mathematical model that combines process planning, cell reconfiguration, task scheduling, and resource allocation, we designed a Genetic Programming Hyper-Heuristic with Bloat Control Mechanism (GPHH-BC) based on multi-heuristic co-evolution. The algorithm employs population segmentation to co-evolve four types of heuristic rules, effectively solving five critical subproblems in dynamic environments while successfully suppressing efficiency degradation caused by rule bloating. Experimental results demonstrate that the proposed method demonstrates a 52.67 % improvement in computational efficiency compared to conventional baseline approaches while ensuring solution feasibility; when compared to state-of-the-art algorithms, it achieves a further 7.40 % improvement in computational efficiency.

Manufacturing scheduling research has often overlooked the complexities of dynamic product assembly and testing scenarios, particularly those involving reconfigurable manufacturing cells (RMCs) and the integration of process planning and scheduling. This article addresses the problem of Dynamic Integrated Process Planning and Scheduling with RMCs, a novel and complex challenge in modern manufacturing systems. A variable-fidelity surrogate-assisted hyper-heuristic algorithm is proposed, which strategically integrates process planning and scheduling tasks to reduce computation time while improving solution quality. Unlike existing methods, our approach uses surrogate models to approximate expensive evaluations, significantly enhancing computational efficiency. In experiments, our method outperformed the second-best approach by 42.4% and the least effective method by 56.6% in terms of computational efficiency, demonstrating its capability to manage dynamic scheduling and cell reconfiguration challenges in large-scale, real-world manufacturing environments.

Many deep learning based railway defect detection methods have been proposed in recent years. They have greatly improved the efficiency and accuracy of defects detection. However, detection of railway defects remains challenging because of the limited number and types of defect samples, thus, general deep learning methods cannot be applied. In this paper, we designed a “Perceptually Constrained Single Image Generative Adversarial Network” (PCSGAN) to expand the number of railway defect image samples. PCSGAN uses a pyramidal structure to learn the internal features of a single image. In addition, we proposed a masking and a perceptual reconstruction loss mechanism to impose specific positional and structural constraints on the images. We tested the method using railway defects images and compared it to other single image generation models. The experiment results show that the images generated by PCSGAN take into account railway prior knowledge, generate railway structure which satisfied the constraints imposed by railway infrastructure designs, and also provide new information. High image realism and lowest Single Image FID were obtained, and the effectiveness of PCSGAN in the defect detection task were also validated.

The wide application of industrial robots has greatly improved assembly efficiency and reliability. However, determining how to efficiently teach a robot to perform assembly manipulation trajectories from demonstration videos is a challenging issue. This paper proposes a method integrating deep learning, image processing, and an iteration model to predict the real assembly manipulation trajectory of a human hand from a video without specific depth information. First, a pose estimation network, Keypoint-RCNN, is used to accurately estimate hand pose in the two-dimensional (2-D) image of each frame in a video. Second, image processing is applied to map the 2-D hand pose estimated by the neural network with the real 3-D assembly space. An iteration model based on the trust region algorithm is proposed to solve for the quaternions and translation vectors of two frames. All the quaternions and translation vectors form the predicted assembly manipulation trajectories. Finally, a UR3 robot is used to imitate the assembly operation based on the predicted manipulation trajectories. The results show that the robot could successfully imitate various operations based on the predicted manipulation trajectories.

Accurate pipe pose estimation plays a pivotal role in the automatic assembly of pipelines. Recently, data-driven deep neural networks have been proven capable of estimating pose. Nonetheless, a large number of labeled datasets are required during the training process. One effective solution is to estimate pose using self-supervised learning. However, existing algorithms are difficult to deal with textureless objects (like pipes), and they avoid the occlusion problem. To this end, in this article, we propose a latent representation self-supervised pose network (LSPN) for accurate monocular pipe pose estimation. We train our network with synthetic RGB (Red, Green, Blue) data, where only a few labeled samples are used to establish the latent pose space, whereas a large number of structured unlabeled samples are used to learn latent pose representation in self-supervised learning. Experiments demonstrate that LSPN achieves excellent performance on real data and is robust to different environments, such as illumination changes and self-occlusion.

Robotic manipulation of deformable linear objects (DLOs) is important in many applications such as the assembly of deformable wire harnesses and cables in manufacturing. Despite some relevant work in modeling, few studies have studied the comprehensive dynamic modeling and precise automatic control of DLOs using robots due to their high degrees of freedom and high flexibility. To fill this gap, the precise single-arm and dual-arm robot manipulation control of DLOs is studied in this article. First, a more comprehensive dynamic model of DLOs is established based on a discrete elastic rod model, which takes into account the twisting deformation of DLOs. The collisions, contacts, and frictions between the DLOs and the plane, as well as the kinematic constraints of both ends of the DLOs, are also considered in the model. Second, practical dynamic control schemes of robots are proposed to realize the precise control of DLOs for both the single-arm control and dual-arm control. For validations, we built an experimental platform using an ABB Yumi robot to implement and validate the proposed approaches in addition to simulations. Finally, various DLO manipulation tasks are conducted and the results for both single-arm and dual-arm manipulations validate the proposed modeling and control approaches.