Due to the increasing demand for agricultural water, the water availability for winter and spring irrigation of cotton fields has decreased. Consequently, dry seeding followed by irrigation (DSSI) has become a widespread cotton cultivation technique in Xinjiang. This study focused on the interaction between soil particles and cotton seeding equipment under DSSI in Xinjiang. The discrete element method (DEM) simulation framework was employed to compare the performance of the Johnson-Kendall-Roberts (JKR) model and Bonding model in simulating contact between soil particles. The models’ ability to simulate the angle of repose was investigated, and shear tests were conducted. The simulation results showed that both models had comparable repose angles, with relative errors of 0.59% for the JKR model and 0.36% for the contact model. However, the contact model demonstrated superior predictive accuracy in simulating direct shear test results, predicting an internal friction angle of 35.8°, with a relative error of 5.8% compared to experimental measurements. In contrast, the JKR model exhibited a larger error. The Bonding model provides a more accurate description of soil particle contact. Subsoiler penetration tests showed that the maximum penetration force was 467.2 N, closely matching the simulation result of 485.3 N, which validates the reliability of the model parameters. The proposed soil simulation framework and calibrated parameters accurately represented soil mechanical properties, providing a robust basis for discrete element modeling and structural optimization of soil-tool interactions in cotton field tillage machinery.

This study investigates a duckbill-type hole seeder to elucidate the kinematic and force characteristics of hole formation under the dry sowing–wet emergence regime and to provide theoretical support for the optimization of key structural parameters. A bidirectional coupling simulation model based on the discrete element method (DEM) and multibody dynamics (MBD) was established to analyze the motion trajectories of the fixed and movable duckbills, the evolution of three-directional forces, and the associated soil–plastic film disturbance under different combinations of front and rear angles. The results indicate that soil disturbance during hole formation is dominated by vertical penetration and uplift, accompanied by forward cutting and lateral redistribution. The three-directional forces acting on the fixed duckbill exhibit a non-monotonic response with respect to the front angle, decreasing first and then increasing, while the force level during the expansion stage of the movable duckbill generally increases with the rear angle. Within the investigated parameter range, a front angle of 18° combined with a rear angle of 38° resulted in a relatively lower overall force level during penetration and expansion, which is favorable for stable hole formation. Field experiments conducted with this configuration showed an average seed placement deviation of 0.50 cm, satisfying the requirements for precision cotton planting under plastic mulch. The findings provide theoretical insight and methodological support for the structural optimization and engineering design of cotton hole seeders operating under the dry sowing–wet emergence regime.

Mulch-film covering technology has been widely adopted in cotton production in arid regions; however, the associated problem of residual-film pollution has become increasingly prominent, creating an urgent demand for efficient and accurate monitoring approaches. Owing to the small target scale, irregular morphology, blurred boundaries, and complex soil backgrounds of residual-film fragments, residual-film detection based on close-range UAV imagery remains a challenging task. To address these issues, this study proposes an improved algorithm, termed YOLO-EGASF, for residual-film detection in arid-region cotton fields, built upon the lightweight YOLOv11n framework. To enhance the detection of small targets with weak boundary characteristics, the baseline model is improved from three aspects. First, a boundary-enhanced multi-branch small-target extraction module (EMSE) is designed to reinforce shallow-layer details and gradient information through multi-scale convolution and explicit edge enhancement. Second, a GLoCA attention module that integrates global coordinate information with local geometric features is constructed to improve the discriminative capability of the model for residual-film targets under complex background conditions. Third, an ASF-layer multi-scale feature fusion structure is introduced, together with an additional small-target detection layer, to strengthen the participation of high-resolution features in cross-scale fusion and prediction. Experimental results on a self-constructed UAV-based residual-film dataset from cotton fields demonstrate that YOLO-EGASF outperforms several mainstream detection models in terms of Precision, Recall, mAP@0.5, and mAP@0.5:0.95, achieving mAP@0.5 and mAP@0.5:0.95 values of 71.9% and 26.8%, respectively. These results indicate a significant improvement in detection accuracy and robustness, confirming that the proposed method can effectively meet the practical requirements of fine-grained residual-film monitoring in arid-region cotton fields.