When uncrewed agricultural machinery performs autonomous operations in the field, it inevitably encounters obstacles such as persons, livestock, poles, and stones. Therefore, accurate recognition of obstacles in the field environment is an essential function. To ensure the safety and enhance the operational efficiency of autonomous farming equipment, this study proposes an improved YOLOv8-based field obstacle detection model, leveraging depth information obtained from binocular cameras for precise obstacle localization. The improved model incorporates the Large Separable Kernel Attention (LSKA) module to enhance the extraction of field obstacle features. Additionally, the use of a Poly Kernel Inception (PKI) Block reduces model size while improving obstacle detection across various scales. An auxiliary detection head is also added to improve accuracy. Combining the improved model with binocular cameras allows for the detection of obstacles and their three-dimensional coordinates. Experimental results demonstrate that the improved model achieves a mean average precision (mAP) of 91.8%, representing a 3.4% improvement over the original model, while reducing floating-point operations to 7.9 G (Giga). The improved model exhibits significant advantages compared to other algorithms. In localization accuracy tests, the maximum average error and relative error in the 2–10 m range for the distance between the camera and five types of obstacles were 0.16 m and 2.26%. These findings confirm that the designed model meets the requirements for obstacle detection and localization in field environments.

Aiming to meet the growing need for automated harvesting, an automatic alignment method based on Euclidean clustering and K-means clustering is proposed to address issues of driver fatigue and inaccurate driving in manually operated corn harvesters. Initially, the corn field environment is scanned using LiDAR to obtain point cloud data, which are then subjected to pass-through filtering and statistical filtering to remove noise and non-corn contour points. Subsequently, Euclidean clustering and K-means clustering methods are applied to the filtered point cloud data. To validate the impact of Euclidean clustering on subsequent clustering, two separate treatments of the obtained point cloud data were conducted during experimental validation: the first used the K-means clustering algorithm directly, while the second involved performing Euclidean clustering followed by K-means clustering. The results demonstrate that the combined method of Euclidean clustering and K-means clustering achieved a success rate of 81.5%, representing a 26.5% improvement over traditional K-means clustering. Additionally, the Rand index increased by 0.575, while accuracy improved by 57% and recall increased by 61%.

Finite element numerical simulations provide a visual and quantitative approach to studying the interaction between rigid mechanical components and flexible agricultural crops. This method is an important tool for the design of modern agricultural production equipment. Obtaining accurate material model parameters for crops is a prerequisite for ensuring the reliability and accuracy of numerical simulations. To address the issue of unclear mechanical constitutive model parameters for industrial hemp stalks, this study utilized the theory of composite materials to establish a mechanical constitutive relationship model for industrial hemp stalks. Compression, tensile, and bending tests on different components of the stalk were conducted, using a computer-controlled universal testing machine, to obtain their elastic parameters. Combined with the measured basic material parameters and contact parameters of industrial hemp stalks, a finite-element numerical simulation model of industrial hemp stalks was established. By conducting Plackett–Burman and central composite experiments, it was determined that among the six measured parameters, the anisotropic plane Poisson’s ratio of the phloem and the isotropic plane Poisson’s ratio of the xylem have a significant influence on the maximum bending force of the stalk. Parameter optimization was carried out, using the relative error of the maximum bending force as the optimization objective, resulting in an anisotropic plane Poisson’s ratio of 0.054 for the phloem and an isotropic plane Poisson’s ratio of 0.28 for the xylem of industrial hemp stalks. To validate the accuracy and reliability of the optimized parameters, a numerical simulation was conducted and compared with the physical experiments. The simulated value obtained was 405.81 N while the actual measured value was 392.55 N. The error between the simulated and measured values was only 3.4%, confirming the effectiveness of the model. The precise parameters for the mechanical characteristics of industrial hemp stalk material obtained in this study can provide a parameter basis for future research on the numerical simulation of mechanized industrial hemp harvesting and retting.

Ramie is an important cash crop in China, and ramie fiber is an important raw material for the textile industry. As a shrub plant, the spatial distribution of the ramie plant is different from that of herbaceous crops, and its plant spacing and row spacing are not fixed, which affects the cutting operation during harvest. In order to solve the above problems, this study constructed a ramie spatial distribution model with statistical methods, and built a prediction model of ramie harvesting feeding quantity on this basis. Based on the analysis of the absolute motion trail of the ramie harvester cutting knife, the calculation equation of the missing cutting area was established, and then the prediction model of the mis-cutting rate was obtained. The results of the ramie field harvest showed that the prediction model of the feed quantity and mis-cutting rate was effective. These methods can provide references to the control and optimization of ramie harvester parameters.

Unmanned agricultural machinery can significantly reduce labor intensity while substantially enhancing operational efficiency and production benefits. However, the presence of various obstacles in complex farmland environments is inevitable. Accurate and efficient obstacle recognition technology, along with a reliable safety warning system, is a crucial prerequisite for ensuring the safe and stable operation of unmanned agricultural machinery. This study proposes a lightweight model for farmland obstacle detection by improving the YOLOv8n object detection algorithm. Specifically, we introduce the Context-Guided Block (CG Block) in the C2f module and the Context-Guide Fusion Module (CGFM) in the Feature Pyramid Network (FPN) to enhance the model’s contextual information perception during feature extraction and fusion. Additionally, we employ a Lightweight Shared Convolutional Separable Batch Normalization Detection Head in the detection head, which significantly reduces the number of parameters while improving detection accuracy. Experimental results demonstrate that our method achieves a mean average precision (mAP) of 92.3% at 59.1 frames per second (FPS). The improved model reduces parameter count and computational complexity by 31.9% and 33.4%, respectively, with a model size of only 4.2 MB. Compared to other algorithms, the proposed model maintains an optimal balance between parameter efficiency, computational cost, detection speed, and accuracy, exhibiting distinct advantages. Furthermore, we propose a safety warning strategy based on the relative velocity and distance between obstacles and the unmanned agricultural machinery. Field experiments conducted under this strategy reveal an overall warning accuracy of up to 86%, verifying the reliability of the safety warning system. This ensures that unmanned agricultural machinery can effectively mitigate potential safety risks during field operations.

Conveying and baling are two important links in the mechanized harvesting of ramie, in the face of ramie cutting and baling harvesting technology research gaps, low stalk conveying rate, high breaking rate and other problems. In this paper, according to the technical requirements of ramie harvesting, we designed a conveying and baling device, the hand-held ramie cutter. First, the key mechanism of the conveying and baling device of the equipment was designed. Then, we analyzed the location of stem clogging and the reasons for the breaking problem during the conveying and baling process. The field harvesting experiments were carried out according to the principles of Box–Behnken experimental design. Taking the machine travelling speed, conveying speed and ramie raking frequency as the test factors and using the Design-Expert V8.0.6.1 to process the data, we established a regression model for each experimental factor on the conveying rate and breaking rate. The order of influence of several factors on the breaking rate is: X₂ > X₁ > X₃; and the effects of the three factors on the conveying rate were X₃ > X₂ > X₁. Through response surface analysis (RSA), the effects of the factors on the indicators were explained, as was the impact of the factors on the indicators. Finally, the parameter optimization was carried out with the delivery rate as the core index. The best combination of motion parameters was obtained as follows: the travelling speed was 0.37 m/s, the chain conveying speed was 1.1 m/s, and the raking frequency was 144 times/min. With the combination of parameters under the field test verification, the results show that compared with the original work quality, the stalk delivery rate increased from 85.2% to 93% (an increase of 7.8%), the stalk breaking rate fell from 31.1% to 20.4% (a decrease of 10.7%). The performance of ramie harvesting and baling was greatly improved, and we achieved relatively satisfactory results.