Search

Air-assisted sprayers are widely used in orchards due to their efficiency in enhancing droplet penetration and deposition. These sprayers disperse droplets through a high-velocity airflow, which agitates the leaves and aids in canopy penetration. This study involved controlled experiments to simulate leaf movement during field spraying, with a focus on the dynamics of peach tree leaves (Prunus persica) in varying wind fields. An experimental setup consisting of a wind-conveying system, a measurement system, and a fixed system was designed. The moving speeds of the wind field (0.75 m/s, 0.5 m/s, and 1.0 m/s) and wind velocities (ranging from 2 m/s to 8 m/s) were varied. Key parameters, including leaf tip displacement, angular velocity, and twisting amplitude, were measured using high-speed cameras and motion analysis software. The results indicate that, at a constant wind velocity, increasing the wind field’s moving speed resulted in a reduced range of motion, decelerated angular velocity, and decreased twisting amplitude of the leaves. Notably, at a wind field speed of 8 m/s and a moving speed of 1.0 m/s, the twisting duration of the leaves was only 67% of that observed at a moving speed of 0.5 m/s. These findings suggest that wind speed and field motion characteristics play a crucial role in leaf dynamics, informing the design of air-assisted spraying systems.

The fruit industry is a significant economic sector in China, with modern orchards gradually transitioning to trellis orchards. For mechanized orchard operations, automatic vehicle navigation is essential. However, in trellis orchards, the shading from trees results in low average light intensity and large variations in lighting, posing challenges for path navigation. To address this, a path navigation algorithm for trellis orchards is proposed based on the UNet-CBAM model. The network structures of UNet, FCN, and SegNet are compared to identify and select the optimal structure for further improvement. Among the three attention mechanisms of channel attention, spatial attention, and combined attention, the most effective mechanism is identified. The optimal attention mechanism is incorporated into the optimized network to enhance the model’s ability to detect path edges and improve detection performance. To validate the effectiveness and generalizability of the model, a total of 400 images were collected under varying lighting intensities. The experimental results show that this method achieves an accuracy of 97.63%, a recall of 93.94%, and an Intersection over Union (IoU) of 92.19%. These results significantly enhance path recognition accuracy in trellis orchards, particularly under low light under conditions. Compared to the FCN and SegNet algorithms, this method provides higher detection accuracy and offers a new theoretical foundation and research approach for path recognition in low-light environments.

Traditional methods of weed control during field management often result in herbicide waste. Precision herbicide application is crucial in agricultural production. This study presents a monodisperse droplet-generation system designed for precision herbicide application, capable of generating monodisperse droplets induced by an electric field. Droplet-generation experiments were conducted to investigate the effects of capillary tube outlet shape, liquid flow rate, and capillary tube size on the generation of charged droplets. A droplet diameter prediction model was established based on the system parameters. Experimental results indicated that as the applied voltage increased, the droplet diameter decreased, and the droplet-generation patterns transitioned sequentially from dripping, micro-dripping, to unstable dripping modes. In a weak electric field, capillaries with beveled outlets produced smaller droplets with more stable diameter distributions compared to those with blunt outlets. In a strong electric field, the smallest droplet diameter from blunt capillaries was 138.2 μm, whereas from beveled capillaries it was 198.7 μm. Within the design parameter range, droplet diameter was basically positively correlated with liquid flow rate and capillary tube size. By controlling the applied voltage, liquid flow rate, and capillary tube size, stable droplet generation could be achieved within a diameter range of 198.7–2520.8 μm, and the coefficient of variation of droplet diameter under the same working conditions was generally less than 6%. The monodisperse droplet-generation system developed in this study can effectively reduce herbicide usage and improve application efficiency.

Comprehensive growth index (CGI) more accurately reflects crop growth conditions than single indicators, which is crucial for precision irrigation, fertilization, and yield prediction. However, many current studies overlook the relationships between different growth parameters and their varying contributions to yield, leading to overlapping information and lower accuracy in monitoring crop growth. Therefore, this study focuses on winter wheat and constructs a comprehensive growth monitoring index (CGIac), based on adaptive weight allocation of growth parameters’ contribution to yield, using data such as leaf area index (LAI), soil plant analysis development (SPAD) values, plant height (PH), biomass (BM), and plant water content (PWC). Using UAV data on vegetation indices, feature selection was performed using the Elastic Net. The growth inversion model was then constructed using machine learning methods, including linear regression (LR), random forest (RF), gradient boosting (GB), and support vector regression (SVR). Based on the optimal growth inversion model for winter wheat, spatial distribution of wheat growth in the study area is obtained. The findings demonstrated that CGIac outperforms CGIav (constructed using equal weighting) and CGIcv (built using the coefficient of variation) in yield correlation and prediction accuracy. Specifically, the yield correlation of CGIac improved by up to 0.76 compared to individual indices, while yield prediction accuracy increased by up to 23.14%. Among the evaluated models, the RF model achieved the best performance, with a coefficient of determination (R2) of 0.895 and a root mean square error (RMSE) of 0.0058. A comparison with wheat orthophotos from the same period confirmed that the inversion results were highly consistent with actual growth conditions in the study area. The proposed method significantly improved the accuracy and applicability of winter wheat growth monitoring, overcoming the limitations of single parameters in growth prediction. Additionally, it provided new technological support and innovative solutions for regional crop monitoring and precision farming operations.

This paper presents the design of an intra-row obstacle avoidance shovel-type weeding machine. Theoretical analysis of intra-row weeding components guided the determination of the structures and parameters for key parts, including the signal acquisition mechanism, automatic obstacle avoidance mechanism, and weeding shovel. Furthermore, a hydraulic system was designed to support these functions. The design aims to optimize intra-row weeding operations, reduce labor costs, enhance weed control effectiveness, and prevent collisions between weeding equipment and grapevines. Through the construction of a mathematical model, the analysis determined the necessary minimum return speed of the hydraulic cylinder for the intra-row weeding shovel to avoid grapevines. We also established a reasonable range for the extension speed of the hydraulic cylinder to minimize areas missed during weeding. Further analysis showed that using the minimum return speed of the hydraulic cylinder effectively reduced missed weeding areas. A virtual prototype model of the weeding machine was created in ADAMS. Using the coverage rate of weeding operation as the evaluation index, single-factor simulation tests determined that the extension speed of the piston rod in the obstacle avoidance hydraulic cylinder and the forward speed of the weeding machine are the main influencing factors. The preset threshold of the control system, which triggered the automatic obstacle avoidance mechanism when the obstacle avoidance rod reached a specific angle (the “Angle Threshold”), was identified as a secondary influencing factor. Other factors were considered irrelevant. Hydraulic cylinder extension speed, weeding machine forward speed, and angle threshold were chosen as the influencing factors. Following the principles of a Box–Behnken experimental design, a quadratic regression combination experiment was designed using a three-factor, three-level response surface analysis method. The evaluation criterion focused on the coverage rate of weeding operation. A regression model was developed to determine the coverage rate of the weeding operation, identifying the optimal parameters as follows: obstacle avoidance hydraulic cylinder extension speed of 120 mm/s, forward speed of the weeding machine at 0.6 m/s, and an angle threshold of 18°. The optimized coverage rate of the weeding operation achieved 86.1%. This study serves as a reference for further optimization of intra-row weeding machines in vineyards and for other crops.

Variable-rate spray systems with canopy leaf area density information detection are an important approach to reducing pesticide usage in orchard management. In order to estimate the canopy leaf area density using ultrasonic sensors, this article proposed three parameter model equations based on ultrasonic echo peaks for canopy leaf area density estimation and verified the accuracy of the three parameter model equations using laboratory-simulated canopy and outdoor tree experiments. The orthogonal regression statistics results from the laboratory-simulated canopy experiment indicated that parameter Vc is more suitable for canopy leaf area density estimation compared to parameter Va and Vb when the density ranges from 0.54 to 5.4 m2m−3. The model equation from parameter Vc has minor systematic errors, and the predicted and observed values of parameter Vc have good agreement with the experimental conditions. The laboratory-simulated canopy and outdoor tree canopy leaf area density verification experiments of parameter Vc were carried out, and the results indicated that the absolute value of the mean relative error is 5.37% in the laboratory-simulated canopy and 2.84% in outdoor tree experiments. The maximum absolute value of the relative error is 8.61% in the laboratory-simulated canopy and 14.71% in the outdoor tree experiments, and the minimum absolute value of the relative error is 3.21% in the laboratory-simulated canopy and 0.56% in the outdoor tree experiments. The laboratory-simulated canopy leaf area density verification results showed that the mean relative errors under canopy leaf area density 0.98 and 4.92 m2m−3 conditions are 6.29% and 5.82%, respectively, which is larger than the mean relative error under 2.95 m2m−3; nevertheless, these results proved that this model equation is applicable for canopy information detection and advanced pesticide application development in future.

To realize the online evaluation of spray quality, an online measuring system for spray coverage and deposition quality is developed. The measuring theory of spray coverage and deposition quality on an LWS (leaf wetness sensor) surface is analyzed. When the spray conditions are constant, there is a linear correlation between the spray coverage on the sensor surface, the spray deposition quality and the LWS output voltage increment. The results of calibration experiments show that when the spray conditions are constant, the coefficient of determination R2 of the regression curves between the sensor output voltage increment and the spray coverage on the sensor surface is more than 0.75, and the coefficient of determination R2 of the regression curves between the sensor output voltage increment and the spray deposition quality on the sensor surface is more than 0.90. Based on ZigBee wireless sensor technology, this paper reports an online measuring system for spray coverage and deposition quality at multiple points in the field. The test results show that the online measuring system has good uniformity. Field test results show that the LWS voltage increment and the coverage rate of water-sensitive paper have a good correlation, and the measuring results of the spray deposition quality trend are in good agreement. The fit of the spray deposition quality curves measured by the two methods was 0.8924. The research results in this paper can provide a reference for using LWS sensors to measure spray coverage and deposition quality.

An air-assisted electrostatic nozzle uses a combination of air-assisted atomization and electrostatic spray technology. This article optimizes the existing air-assisted electrostatic nozzles in terms of structural design to obtain a higher charge-to-mass ratio and a smaller droplet size. The optimized air-assisted electrostatic nozzle was studied experimentally, and the effects of liquid pressure, air pressure and applied voltage on the droplet size and charge-to-mass ratio were investigated. Comparing the effects of air pressure, liquid pressure and applied voltage on the charge-to-mass ratio and droplet size, the relationship curves of the droplet size and charge-to-mass ratio under each voltage were fitted using the Rayleigh charge limit theory. For a higher CMR during the spray operation, applied voltages between 2.5 kV and 3 kV, an air pressure between 0.4 bar and 0.6 bar, and a liquid pressure of less than 0.9 bar could be chosen. The optimized air-assisted electrostatic nozzles not only have small droplets but also have high charge-to-mass ratios, reducing the need for pesticide use and thus protecting human health and the environment.

In the process of applying plant protection sprays, the atomisation process of complex pesticide components such as emulsion pesticides is different from that of water. Indeed, emulsion is often used as an additive to spray to reduce drift. Therefore, this study investigated the different morphological characteristics that occur between emulsions and water during atomisation at different pressures through visualisation experiments and interpreting the formation of structural differences between the two fragmentation mechanisms. The effect of liquid sheet structure on droplet size distribution was analysed in three-dimensional space, not only from one spatial perspective, but how it alters the morphological structures of liquid sheet leading to different potential droplet drift characteristics. It was found that the smaller the liquid sheet disturbance, the more concentrated the droplet size distribution, the more intense the liquid sheet disturbance, the more dispersed the droplet size distribution. The addition of 0.02% emulsion significantly reduced the proportion of V100 (the ratio of volume with drops smaller than 100 μm to the total volume of all droplets) from 21.33% to 10.24%, and the higher the emulsion concentration, the smaller the V100. The ability of the emulsion to increase V400 decreased with increasing pressure.

The accurate detection of canopy characteristics is the basis of precise variable spraying. Canopy characteristics such as canopy density, thickness and volume are needed to vary the pesticide application rate and adjust the spray flow rate and air supply volume. Canopy thickness is an important canopy dimension for the calculation of tree canopy volume in pesticide variable spraying. With regard to the phenomenon of ultrasonic waves with multiple reflections and the further analysis of echo signals, we found that there is a proportional relationship between the canopy thickness and echo interval time. In this paper, we propose a method to calculate canopy thickness using echo signals that come from ultrasonic sensors. To investigate the application of this method, we conducted a set of lab-based experiments with a simulated canopy. The results show that we can accurately estimate canopy thickness when the detection distance, canopy density, and canopy thickness range between 0.5and 1.5 m, 1.2 and 1.4, and 0.3and 0.6 m, respectively. The relative error between the estimated value and actual value of the simulated canopy thickness is no higher than 8.8%. To compare our lab results with trees in the field, we measured canopy thickness from three naturally occurring Osmanthus trees (Osmanthus fragrans Lour). The results showed that the mean relative errors of three Osmanthus trees are 19.2%, 19.4% and 18.8%, respectively. These results can be used to improve measurements for agricultural production that includes both orchards and facilities by providing a reference point for the precise application of variable spraying.