Conventional continuous pesticide application remains prevalent in agriculture, but its limitations in addressing the spatial–temporal variability of biotic stressors have led to excessive chemical inputs and inefficiency. The emergence of precision agriculture has catalyzed significant advancements in variable-rate spray systems to optimize agrochemical deployment through real-time modulation. This technology demonstrates critical advantages in minimizing the environmental footprint while maintaining crop protection efficacy. Our systematic review analyzes three foundational variable-rate spray architectures—pressure-regulated, flow rate-regulated, and pesticide concentration-regulated mechanisms—evaluating their maturity and implementation paradigms. Pressure-regulated technology relies on the pressure–flow relationship to achieve regulation, but there is a narrow range in flow regulation, atomization stability is insufficient, and there are other defects. Flow rate-regulated technology achieves precise control through the dynamic adjustment of the nozzle orifice area or Pulse-Width Modulation duty cycles, but this technology faces mechanical wear, a nonlinear flow–duty cycle relationship, and other challenges. Pesticide concentration-regulated technology is centered on real-time mixing, which can avoid the residue of chemicals but is highly dependent on fluid characteristics and mixing efficiency. This study proposes improvement paths from the perspectives of hardware optimization, control strategy integration, and material innovation. Through the summary and analysis of this paper, we hope to provide valuable references for future research on variable-rate spray technology.

Accurate crop row detection is an important foundation for agricultural machinery to realize autonomous operation. Existing methods often compromise between real-time performance and detection accuracy, limiting their practical field applicability. This study develops a high-precision, efficient crop row detection algorithm specifically optimized for soybean–corn compound planting conditions, addressing both computational efficiency and recognition accuracy. In this paper, a real-time soybean–corn crop row detection method based on GD-YOLOv10n-seg with principal component analysis (PCA) fitting was proposed. Firstly, the dataset of soybean–corn seedling crop rows was established, and the images were labeled with line labels. Then, an improved model GD-YOLOv10n-seg model was constructed by integrating GhostModule and DynamicConv into the YOLOv10n-segmentation model. The experimental results showed that the improved model performed better in MPA and MIoU, and the model size was reduced by 18.3%. The crop row center lines of the segmentation results were fitted by PCA, where the fitting accuracy reached 95.08%, the angle deviation was 1.75°, and the overall processing speed was 61.47 FPS. This study can provide an efficient and reliable solution for agricultural autonomous navigation operations such as weeding and pesticide application under a soybean–corn compound planting mode.

In this article, we propose a fuzzy proportional–integral–derivative (Fuzzy-PID) controller that integrates a system-identification-based control strategy. We aim to address the challenge of regulating electrical conductivity (EC) in a fertigation system to ensure precise nutrient delivery. During fertilization, the nutrient solution EC value increases gradually and nonlinearly as water and fertilizer are integrated. Precise fertilizer injection is essential to maintain stable EC levels, preventing crop undernutrition or overnutrition. The fertigation process is modeled using a particle swarm optimization (PSO)-based system identification method. A Fuzzy-PID method is then employed to regulate the nutrient solution EC value based on the pre-determined or real-time identified transfer model. The proposed control strategy is deployed within a programmable logic controller (PLC) environment and validated on a PLC-based fertilizer system. The results show that the identified transfer model accurately represents the fertilizer mixing process, achieving a standard Mean Absolute Percentage Error (MAPE) value of less than 5% within 2 s using the proposed PSO-based identification method. In the simulation tests, the proposed Fuzzy-PID control rule would converge the nutrient solution to target EC values 1000 and 1500 μs/cm within a deviation band ± 50 μs/cm, within 6 s from the recorded identified transfer models and within 25 s from the real-time identified transfer models. In the device’s test, the convergence time of the fertigation EC control is approximately 16 s from the history data and 42 s from the real-time collected data, with a deviation band ± 50 μs/cm. In contrast, it may take over 70 s for the EC regulation of the same fertilization, using the classic control methods including conventional PID and Fuzzy-PID. The proposed control strategy significantly improves EC regulation in terms of speed, stability, and precision, enhancing the performance of fertilizer mixing systems.

Although there are existing interplant weed control devices for soybeans, they mostly rely on image recognition and intelligent navigation platforms. Simultaneously, automated weed control devices are not yet fully mature, resulting in issues such as high seedling injury rates and low weeding rates. This paper proposed a reciprocating interplant weed control device for soybeans based on the idea of intermittent reciprocating opening and closing of weeding execution components. The device consists of a laser ranging sensor, servo motor, Programmable Logic Controller (PLC), and weeding mechanism. Firstly, this paper explained the overall structure and working principle of the weed control device, and discussed the theoretical analysis and structural design of the critical component, elastic comb teeth. This paper also analyzed the working principle of the elastic comb teeth movement trajectory and seedling avoidance action according to soybean agronomic planting requirements. Then, field experiments were conducted, and the experiment was designed by the quadratic regression general rotation combination experimental method. The number of combs, the speed of the field management robot, and the stabbing depth were taken as the test factors to investigate their effects on the test indexes of weeding rate and seedling injury rate. The experiment utilized a response surface analysis method and designed a three-factor, three-level quadratic regression general rotation combination experimental method. The results demonstrate that the number of comb teeth has the most significant impact on the weeding rate, while the forward speed has the most significant impact on the seedling injury rate. The optimal combination of 29.06 mm stabbing depth, five comb teeth, and a forward speed of 0.31 m/s achieves an optimal operational weeding rate of 98.2% and a seedling injury rate of 1.69%. Under the optimal parameter combination conditions, the machine’s performance can meet the requirements of intra-row weeding operations in soybean fields, and the research results can provide a reference for the design and optimization of mechanical weed control devices for soybean fields.

The high cost of manual weed control and the overuse of herbicides restrict the yield and quality of soybean. Intelligent mechanical weeding and precise application of pesticides can be used as effective alternatives for weed control in the field, and these require accurate distinction between crops and weeds. In this paper, images of soybean seedlings and weeds in different growth areas are used as datasets. In the aspect of soybean recognition, this paper designs a YOLOv8nGP algorithm with a backbone network optimisation based on GhostNet and an unconstrained pruning method with a 60% pruning rate. Compared with the original YOLOv8n, the YOLOv8nGP improves the Precision (P), Recall (R), and F1 metrics by 1.1% each, reduces the model size by 3.6 mb, and the inference time was 2.2 ms, which could meet the real-time requirements of field operations. In terms of weed recognition, this study utilises an image segmentation method based on the Normalized Excess Green Index (NExG). After filtering the soybean seedlings, the green parts of the image are extracted for weed recognition, which reduces the dependence on the diversity of the weed datasets. This study combines deep learning with traditional algorithms, which provides a new solution for weed recognition of soybean seedlings.

The traditional continuous, quantitative spraying technology ignores the severity of pests, diseases and grasses, spatial distribution and other differences, resulting in low effective utilization of pesticides, environmental pollution and other problems. Variable-rate spray technology has become an important development direction in the field of precision agriculture by dynamically sensing crop canopy morphology, pest and disease distribution, and environmental parameters, adjusting the application amount in real time, and significantly improving pesticide utilization. In this study, we systematically review the core progress of variable-rate spray technology; focus on the technical system of information detection, spray volume model, and control system; analyze the current bottlenecks; and propose an optimization path to adapt to the complex agricultural conditions. At the level of information perception, LiDAR, machine vision, and multi-source sensor fusion technology constitute the main perception architecture, and infrared and ultrasonic sensors assist target recognition in complex scenes. In the construction of the spray volume model, models based on canopy volume, leaf area density, etc., are used to realize dynamic application decision by fusing equipment operating parameters, pest and disease levels, meteorological conditions, and so on. The control system takes the solenoid valve + PID control as the core program, and improves the response speed through PWM regulation and closed-loop feedback. The current technical bottlenecks are mainly concentrated in the sensor dynamic detection accuracy, model environmental adaptability, and the reliability of the execution parts. In the future, it is necessary to further promote anti-jamming multi-source heterogeneous sensor data fusion, multi-factor adaptive spray model development, lightweight edge computing deployment, and solenoid valve structural parameter optimization and other technical research, with a view to promoting the application of variable-rate spray technology to the field on a large scale and providing a theoretical reference and technological support for the green transformation of agriculture.