Accurate crop row detection is identified as a critical prerequisite for autonomous agricultural navigation, yet it remains challenging in complex field environments. To achieve a balance between segmentation accuracy, robustness, and real-time performance, an improved crop row segmentation and navigation method based on the DeepLabV3+ framework was developed. MobileNetV2 was adopted as the backbone to minimize computational costs, while feature representation was enhanced through integrated attention mechanisms and multi-scale fusion. Specifically, split-attention convolution was integrated into the backbone, a DenseASPP + SP module was employed for multi-scale contextual capture, and a Convolutional Block Attention Module (CBAM) was added to refine feature responses. Experimental results demonstrated that the proposed method outperformed mainstream models, achieving a mean Intersection over Union (mIoU) of 93.42% and an f1-score of 96.8%. The model maintained a lightweight architecture with 8.35 M parameters and a real-time speed of 32 FPS. Furthermore, crop row anchor points were extracted and processed via DBSCAN clustering and RANSAC fitting to generate high-precision navigation lines. Validation showed that the middle crop row yielded the highest fitting accuracy with minimal angular and lateral errors. This study provides an efficient visual perception solution for intelligent field operations.

Chinese Assembled Solar Greenhouses (CASGs) in the Gobi Desert region face significant diurnal temperature variations, with excessively high temperatures during the day and low temperatures at night, which adversely affect crop growth. Traditional temperature regulation technologies are hindered by high energy consumption, high costs, and severe pollutants. To address these issues, this study designed a heating system suitable for CASGs in the Gobi Desert region, integrating solar air source heat pump technology with underground pipe systems. The power consumption and performance of the system were assessed by comparing temperature and humidity in an experimental greenhouse (with the system), a control greenhouse (without the system), and outdoor environments under various typical climate conditions. The results indicated that the system exhibited excellent performance in both daytime heat absorption and nighttime heat release. Specifically, during operation, the maximum daytime temperature in the experimental greenhouse was reduced by up to 5 °C, while the minimum nighttime temperature increased by up to 8 °C, effectively preventing crop frost damage. The system achieved heat absorption rates of 14 to 16 KJ s⁻¹ and heat release rates of 36.5 to 37.5 KJ s⁻¹, with average coefficients of performance (COP) of 4.33 and 4.81. Compared to traditional heating methods using coal, gas, and electricity, the system reduced energy consumption by 84.7%, 81.3%, and 79.1%, respectively, and decreased greenhouse gas emissions by 8.24 t, 6.52 t, and 5.67 t, respectively. This system exhibits outstanding thermal efficiency, energy savings, and environmental benefits, while also showing promising economic benefits with a payback period of four years, providing a reliable heating solution for CASGs in the Gobi Desert region.