Robotic positioning accuracy is critically affected by both geometric and non-geometric errors. To address this dual error issue comprehensively, this paper proposes a novel two-stage compensation framework. First, a Memory based red billed blue magpie optimizer (MRBMO) is employed to identify and compensate for geometric errors by optimizing the geometric parameters based on end-effector observations. This memory-guided evolutionary mechanism effectively enhances the convergence accuracy and stability of the geometric calibration process. Second, a tuned graph neural network (AMPSO-GNN) is developed to model and compensate for non-geometric errors, such as cable deformation, thermal drift, and control imperfections. The GNN architecture captures the topological structure of the robotic system, while the adaptive momentum PSO dynamically optimizes the network’s hyperparameters for improved generalization. Experimental results on a six-axis industrial robot demonstrate that the proposed method significantly reduces residual positioning errors, achieving higher accuracy compared to conventional calibration and compensation strategies. This dual-compensation approach offers a scalable and robust solution for precision-critical robotic applications.

As a key indicator of structural integrity and in-service performance, crack detection is essential for the condition assessment and preventive maintenance of bridges. To address the challenges of detecting cracks with various scales and shapes under low-contrast backgrounds in bridge inspection tasks, this paper proposes a robust detection method named CLGDS. It is based on YOLO11 with enhanced feature fusion and SIoU loss optimization, which effectively improves the accuracy and robustness of crack identification. The proposed framework includes three key innovations. (1) A Cross Stage Partially Large Separable Kernel Attention (C2LSKA) module is integrated in the backbone network to enhanced the representation of crack features in the case of morphologically diverse and complex background interference. (2) A Gathering and Distributing (GD) mechanism serves as the neck network, facilitating multi-scale feature fusion and improving the detection performance for cracks of varying scales and geometrically irregular edges. (3) A Scylla-IoU (SIoU) loss function is introduced to replace the commonly used Complete IoU (CIoU) loss. By explicitly incorporating directional sensitivity and multi-scale adaptability, SIoU effectively mitigates angle-dependent misalignment during bounding box regression. Experimental results demonstrate that CLGDS achieves a mean average precision (mAP@50) of 93.5%, outperforming YOLOv5, YOLOv8, and YOLO11 by margins of +1.5%, +0.6%, and +1.2%, respectively. Furthermore, it attains a mAP@50-95 of 68.5%, significantly higher than that of YOLOv5 (61.3%), YOLOv8 (62.6%), and YOLO11 (65.3%). These results validate the effectiveness of CLGDS in accurate bridge crack detection, providing a solid technical foundation for automated structural health monitoring and preventive maintenance.

As an alternative to true isotropic 3D imaging, image super-resolution (SR) has been applied to reconstruct an isotropic 3D volume from multiple anisotropic scans. However, traditional SR methods struggle with inadequate performance, prolonged processing times, and the necessity for manual feature extraction. Motivated by the exceptional representational ability and automatic feature extraction of convolutional neural networks (CNNs), in this work, we present an end-to-end isotropic MRI reconstruction strategy based on deep learning. The proposed method is based on 3D convolutional neural networks (3D CNNs), which can effectively capture the 3D structural features of MRI volumes and accurately predict potential structure. In addition, the proposed method takes multiple orthogonal scans as input and thus enables the model to use more complementary information from different dimensions for precise inference. Experimental results show that the proposed algorithm achieves promising performance in terms of both quantitative and qualitative assessments. In addition, it can process a 3D volume with a size of 256 × 256 × 256 in less than 1 min with the support of an NVIDIA GeForce GTX 1080Ti GPU, which suggests that it is not only a quantitatively superior method but also a practical one.

To address the issues of parameter redundancy and low computational efficiency in traditional convolutional neural networks (CNNs) for motor bearing fault diagnosis, which are caused by increasing network depth, this paper proposes a Disentangle-and-Aggregate Feature Learning Network (DAFNet). This method is designed to overcome the challenges of model deployment on resource-constrained edge devices. Moving beyond the conventional strategy of simply stacking layers, DAFNet innovatively adopts a hierarchical disentanglement and aggregation mechanism. It utilizes a secondary splitting strategy to disentangle shallow, medium, and deep features, followed by terminal feature fusion to achieve an effective representation of fault information. Experimental results based on the CWRU dataset demonstrate that DAFNet achieves a 100% average accuracy in fault diagnosis while significantly reducing both computational overhead and parameter count. Compared with existing mainstream lightweight models, this method exhibits superior generalization performance and inference speed, providing new theoretical support for the efficient application of deep learning in industrial embedded systems.

Accurate absolute positioning is essential for industrial robot arms, especially in high-precision manufacturing tasks. Traditional calibration methods often rely heavily on domain-specific knowledge and handcrafted algorithms, making it challenging for broader adoption across disciplines. To tackle this problem, this paper proposes a novel calibration framework based on an enhanced metaheuristic approach named RLSGA, which integrates deep reinforcement learning with the Snow Geese Algorithm (SGA). Unlike conventional strategies where the movement of agents is fully determined by predefined equations, the proposed method leverages a deep policy network to guide individual geese’s migration behavior. This network generates adaptive decisions regarding position updates, convergence direction, and flight mode selection. The learned policy enables more flexible and efficient exploration of the calibration parameter space. Experimental results on robot arm calibration tasks demonstrate that RLSGA achieves superior calibration accuracy and robustness compared to existing optimization-based methods, validating its effectiveness and potential for real-world applications.