Due to the inherent limitations of detection networks and the imbalance in training data, small-target detection has always been a challenging issue in the field of target detection. To address the issues of false positives and missed detections in small-target detection scenarios, a new algorithm based on STDA-YOLOv8 is proposed for small-target detection. A novel network architecture for small-target detection is designed, incorporating a Contextual Augmentation Module (CAM) and a Feature Refinement Module (FRM) to enhance the detection performance for small targets. The CAM introduces multi-scale dilated convolutions, where convolutional kernels with different dilation rates capture contextual information from various receptive fields, enabling more accurate extraction of small-target features. The FRM performs adaptive feature fusion in both channel and spatial dimensions, significantly improving the detection precision for small targets. Addressing the issue of a significant disparity in the number of annotations between small and larger objects in existing classic public datasets, a new data augmentation method called Copy–Reduce–Paste is introduced. Ablation and comparative experiments conducted on the proposed STDA-YOLOv8 model demonstrate that on the VisDrone dataset, its accuracy improved by 5.3% compared to YOLOv8, reaching 93.5%; on the PASCAL VOC dataset, its accuracy increased by 5.7% compared to YOLOv8, achieving 94.2%, outperforming current mainstream target detection models and small-target detection algorithms like QueryDet, effectively enhancing small-target detection capabilities.

The A* algorithm is widely used in mobile robot path planning; however, it faces challenges such as unsmooth planned paths, redundant nodes, and extensive search areas. This paper proposes a hybrid algorithm combining an improved A* algorithm with the Dynamic Window Approach. By quantifying grid obstacle data to extract environmental information and employing a grid-based environmental modeling method, the proposed approach enhances path smoothness at turns using second-order Bezier curve smoothing. It improves the heuristic function and child node selection process, applying these advancements in experimental path planning scenarios. A simulated 2D map was constructed using point cloud scanning in RViz to validate the hybrid algorithm through simulations and real-world outdoor tests. Experimental results demonstrate that, compared to the A* and DWA algorithms, the improved hybrid algorithm enhances search efficiency by 10.93%, reduces search node count by 32.26%, decreases the number of turning points by 36.36% and the value of turning angle by 34.83%, shortens the total path length by 22.05%, and improves overall path smoothness. Simulations and field tests confirm that the proposed hybrid algorithm is more stable, significantly reduces collision probability, and demonstrates its applicability for mobile robot localization and navigation in real-world environments.