Object detection is becoming increasingly critical in autonomous driving. However, the accuracy and effectiveness of object detectors are often constrained by the obscuration of object features and details in adverse weather conditions. Therefore, this paper presented the DAN-YOLO vehicle object detector specifically designed for driving conditions in adverse weather. Building on the YOLOv7-Tiny network, SPP was replaced with SPPF, resulting in the SPPFCSPC structure, which enhances processing speed. The concept of Hybrid Dilated Convolution (HDC) was also introduced to improve the SPPFCSPC and ELAN-T structures, expanding the network’s receptive field (RF) while maintaining a lightweight design. Furthermore, an efficient multi-scale attention (EMA) mechanism was introduced to enhance the effectiveness of feature fusion. Finally, the Wise-IoUv1 loss function was employed as a replacement for CIoU to enhance the localization accuracy of the bounding box (bbox) and the convergence speed of the model. With an input size of 640 × 640, the DAN-YOLO algorithm proposed in this study achieved an increase in mAP0.5 values of 3.4% and 6.3% compared to the YOLOv7-Tiny algorithm in the BDD100K and DAWN benchmark tests, respectively, while achieving real-time detection (142.86 FPS). When compared with other state-of-the-art detectors, it reports better trade-off in terms of detection accuracy and speed under adverse driving conditions, indicating the suitability for autonomous driving applications.

With the development of intelligent manufacturing and smart logistics, automated guided vehicles (AGVs) are gradually expanding from traditional indoor environments to factory roads, logistics parks, industrial parks, and other outdoor road environments. However, low-visibility conditions, such as fog, rain, snow, and sandstorms, can degrade image quality, weaken object boundaries, and reduce the accuracy of visual perception systems. To address these challenges, an improved lightweight object detection model based on YOLOv11, named DMSM-YOLO, is proposed. To handle blurred boundaries, weak target features, and background interference in low-visibility images, the model is optimized from feature extraction, feature fusion, and prediction refinement. First, EPConv is designed to enhance directional contour interaction and improve blurred-boundary representation. Second, C3k2-MDFI is constructed to strengthen multi-scale weak-target representation. Third, MLCA is introduced to recalibrate low-contrast fused features by combining local spatial details and global contextual information. Finally, SMDetect is developed to improve localization accuracy and confidence estimation for blurred and occluded objects. Experimental results show that the proposed model achieves mAP@0.5 values of 54.71% and 71.67% on the DAWN and RTTS datasets, respectively. Compared with YOLOv11n, the mAP@0.5 is improved by 5.36 and 3.26 percentage points, respectively, while the model maintains a compact size of 2.56 M parameters and 7.1 GFLOPs. Additional experiments further verify the effectiveness of the proposed method under low-light conditions. The proposed model improves detection accuracy and robustness under low-visibility conditions while maintaining lightweight characteristics, demonstrating its potential for AGV visual perception.