Xiaoxi Hao

Rice leaf disease poses a significant threat to global food security and ecological stability. While existing studies predominantly concentrate on detecting symptoms after visible lesions emerge, early-stage disease features—which are often subtle—are critical for timely intervention. This paper introduces a novel detection approach based on an enhanced YOLOv11n architecture, tailored for the precise and efficient recognition of early-stage rice leaf disease indicators. To address the limitations of traditional detection techniques in identifying fine-grained features, we propose three key modules: the Multi-branch Large-kernel Fusion Depthwise (MLFD) module, the Multi-scale Dilated Transformer-based Attention (MDTA) module, and the Lightweight Detection Head (Lo-Head). The MLFD module enhances multi-scale feature extraction via parallel pathways and depthwise convolutions with large kernels. The MDTA module integrates both spatial and channel attention through a multi-head mechanism, improving the representation of diverse lesion features. Meanwhile, the Lo-Head detection head significantly reduces model complexity and parameter count, facilitating deployment on edge devices without compromising accuracy. Experimental results show that the proposed network achieves substantial performance gains. At an input resolution of 640×640, the model reaches a mean Average Precision (mAP@50:95) of 0.7927—an increase of 1.84 percentage points over the baseline YOLOv11n. It also outperforms Faster R-CNN, YOLOv5n, YOLOv8n, and YOLOv10n by 17%, 7.2%, 3%, and 2.5% respectively, while maintaining a low computational load of 6.2 GFLOPs and 2.66M parameters. These findings underscore the model’s potential for real-world agricultural applications, particularly in enabling early detection and precise disease control. The proposed method represents a step toward proactive plant health monitoring and precision agriculture.

To solve the complicated parameter tunning problem of bridge crane sliding mode controller, a sliding mode controller based on IGWO (Improved Grey Wolf Algorithm) was designed. Different from the existing methods, the proposed method can obtain good control effect without going through the complicated manual adjustment process. Specifically, a new switching function is designed to reduce chattering. In addition, the population diversity of traditional gray Wolf algorithms is poor. To enhance its search ability, an IGWO algorithm based on neighborhood learning is proposed and the controller parameters are set with it. The simulation consequences show that the proposed controller can position and anti-swing well.

During the transference of a ladle by the casting crane, the antiswing control of the ladle is particularly difficult due to liquid sloshing, so we designed a model based on a radial spring damper. During the ladle swings, the centrifugal force causes the spring damper to move radially, thereby generating a Coriolis force that inhibits the sloshing of the liquid. In addition, a sloshing analysis of the liquid in the ladle is carried out, and a double-pendulum casting crane model based on viscous damping and radial spring damper is established. On the basis of this model, the Enhanced Coupled Adaptive Sliding Mode Control (ECASMC) method is proposed. By introducing an enhanced coupling variable and constructing a new coupling deviation signal, we enhance the relationship among state quantities. Then, a new type of sliding surface is designed based on the enhanced coupling deviation signal, and adaptive technology is used to adjust the sliding mode parameters, which enhances the system's robustness for system parameter variations and external disturbances. Using LaSalle's invariance principle and Lyapunov theorem, we prove that the casting crane system is asymptotically stable near the desired equilibrium point. The simulation results verify the effectiveness and superior control performance of the proposed method even in the presence of uncertainties.