To satisfy the obstacle avoidance requirements of unmanned agricultural machinery during autonomous operation and address the challenge of rapid obstacle detection in complex field environments, an improved field obstacle detection model based on YOLOv8 was proposed. This model enabled the fast detection and recognition of obstacles such as people, tractors, and electric power pylons in the field. This detection model was built upon the YOLOv8 architecture with three main improvements. First, to adapt to different tasks and complex environments in the field, improve the sensitivity of the detector to various target sizes and positions, and enhance detection accuracy, the CBAM (Convolutional Block Attention Module) was integrated into the backbone layer of the benchmark model. Secondly, a BiFPN (Bi-directional Feature Pyramid Network) architecture took the place of the original PANet to enhance the fusion of features across multiple scales, thereby increasing the model’s capacity to distinguish between the background and obstacles. Third, WIoU v3 (Wise Intersection over Union v3) optimized the target boundary loss function, assigning greater focus to medium-quality anchor boxes and enhancing the detector’s overall performance. A dataset comprising 5963 images of people, electric power pylons, telegraph poles, tractors, and harvesters in a farmland environment was constructed. The training set comprised 4771 images, while the validation and test sets each consisted of 596 images. The results from the experiments indicated that the enhanced model attained precision, recall, and average precision scores of 85.5%, 75.1%, and 82.5%, respectively, on the custom dataset. This reflected increases of 1.3, 1.2, and 1.9 percentage points when compared to the baseline YOLOv8 model. Furthermore, the model reached 52 detection frames per second, thereby significantly enhancing the detection performance for common obstacles in the field. The model enhanced by the previously mentioned techniques guarantees a high level of detection accuracy while meeting the criteria for real-time obstacle identification in unmanned agricultural equipment during fieldwork.

In this study, a 4UM-120D electric leafy vegetable harvester was employed as the research object. An automatic control system was created to maintain the cutter’s height above the ground within ±2% of the desired value. The intention was to reduce the operators’ work intensity while improving the leafy vegetable harvester’s working quality. The automatic control system for the cutter height from the ground was explained, along with its structure and operating philosophy. MATLAB was used to establish the two-phase hybrid stepper motor’s mathematical electrical equation and mechanical equation models. An analysis was carried out on the fundamentals and differences between position PID and incremental PID control algorithms. Utilizing incremental PID in combination, the control strategy for the harvester cutter height from the ground was built, and an automatic control system was produced under the corresponding control strategy. The stability, accuracy, and rapidity of the automatic control system of the cutter height from the ground under the incremental PID control strategy were analyzed by simulating different actual working conditions with MATLAB/Simulink and taking the steady-state transition time as the evaluation index. The test results show that when the deviation between the current value and the set value was greater than 2%—that is, when the harvester was in the condition of suddenly crossing the ditch or suddenly climbing the slope—the automatic control system based on the incremental PID control strategy had a good dynamic response performance and stability. This resulted in the automatic control function of the harvester cutter height off the ground being achieved. When the rotation angle PID control algorithm’s proportional coefficient is Kp = 4.665, the rotation speed PID control algorithm’s proportional coefficient is Kp = 5.65 and its integral coefficient is Ki = 3.86, and the current PID control algorithm’s proportional coefficient is Kp = 0.5455 and its integral coefficient is Ki = 30.4578. The harvester abruptly crossed a ditch while operating steadily, and the automatic control system’s steady-state transition time for the height of the cutter off the ground was 1.0811 s. The harvester abruptly climbed a slope while operating steadily, and the automatic control system’s steady-state transition time for the height of the cutter off the ground was 1.1185 s. Data from the field tests revealed a degree of reliability in the simulation test results. The study offered a strategy for raising the harvester quality for leafy vegetables while lowering the operator workload.

This study proposes a variable-scale optimization strategy for land-levelling path planning to overcome the limitations of conventional traversal-based operations, including poor coordination, insufficient planning, low operational efficiency, and the computational burden associated with large datasets and constrained earthmoving capacity. For large-scale inter-regional earthwork balancing, an improved ant colony optimization (IACO) algorithm is developed to generate efficient region to region transfer routes. After verifying that inter-regional earthwork balance satisfies the levelling requirement, a field-wide fine-levelling plan is produced at the grid scale using a hybrid method that integrates an improved A* search with ant colony optimization (FIA*ACO). The proposed framework is evaluated through simulation and field experiments using measurement-based indicators, including the maximum elevation difference and the proportion of points within ±5 cm of the target elevation. Field results show that IACO-based inter-regional planning increases the ±5 cm compliant proportion by 14.18 percentage points and reduces the maximum elevation difference by 0.079 m. Subsequent FIA*ACO-based fine-gridded planning further improves the ±5 cm compliant proportion by 20.82 percentage points and decreases the maximum elevation difference by 0.311 m. Overall, the results demonstrate that inter-regional planning rapidly expands the area meeting levelling standards, while grid-level refinement further enhances levelling quality, validating the effectiveness of the proposed variable-scale strategy for land-levelling path planning.

The study developed a PID-based automatic control system for travel speed based on the analysis of the entire machine structure of the 4UM-120D electric leafy vegetable harvester to enhance the operation quality of the harvester and lessen the workload of operators. Taking the system adjustment time during the process of starting and maintaining the travel speed of the harvester within ±2% of the set value as the main evaluation index, the Box–Behnken test method was applied based on the single-factor test, and the proportional coefficient Kp, the integral coefficient Ki and the initial value U of the travel motor drive voltage of the PID control algorithm were used as the test factors, and the PID-based 4UM-120D electric leafy vegetable harvester travel speed automatic control system operating parameters based on PID was studied in a three-factor, three-level experiment, and a multiple regression model of evaluation indexes on each factor was established to analyze the influence of each factor on system rapidity and to obtain the optimal operating parameters. The test findings demonstrated that the beginning value of the driving voltage of the travel motor U, the integration factor Ki, and the proportionality factor Kp had the largest to smallest influences on the system regulation time. When the initial value of the driving voltage of the travel motor was 1.81 V, the integration factor was 0.020, the proportionality factor was 0.127, the system regulation time was 2.14 s, and the relative error was 1.8% compared with the optimized value. The results of the study could provide a way to improve the operation quality of leafy vegetable harvesters and reduce the work intensity of operators.

This study designed a control system for the automatic unloading and replenishment of baskets based on the cooperative detection of photoelectric sensors and pressure sensors based on analyzing the structure of the 4UM-120D electric leafy vegetable harvester. The goal of this study was to increase the operation efficiency of leafy vegetable harvesters and decrease the work intensity of operators. A control system for the automatic unloading and replenishment of baskets based on the cooperative detection of a photoelectric sensor and pressure sensor was designed and constructed after an analysis of the operating principle and system components of automatic basket unloading and basket replenishment control at the rear of the harvester. The bench test results showed that the bottom photoelectric sensor and top photoelectric sensors 1 and 2 on the touch screen were not lit and the pressure sensor value was displayed as −0.00075531 kg, after pressing the system start button on the touch screen. On the touch screen, only the basket feeding motor was on: the transverse conveyor motor and the basket unloading motor were not, indicating that there was no collection basket on the unloading basket conveyor belt at this time and that the basket feeding motor was conveying an empty basket to the unloading basket conveyor belt. At 26 s, on the touch screen, only the top photoelectric sensor 2 was not on: the top photoelectric sensor 1 and the bottom photoelectric sensor were on and the pressure sensor value was shown as 1.38488 kg. Only the transverse conveyor motor lit up on the touch screen, the basket unloading motor and the basket feeding motor did not light up, indicating that the leafy vegetables temporarily stored in the transverse conveyor belt started to fall into the collection basket at this time and had not yet reached the expected capacity of the collection basket. At 43 s, the bottom photoelectric sensor and top photoelectric sensors 1 and 2 were lit on the touch screen and the pressure sensor value was shown as 2.37229 kg. On the touch screen, only the basket unloading motor lit up: the transverse conveyor motor and the basket feeding motor were not lit up, indicating that the collection basket capacity had reached the expected capacity at this time and the unloading was in progress. At 83 s, the bottom photoelectric sensor and top photoelectric sensors 1 and 2 were not lit on the touch screen and the pressure sensor value was displayed as −0.0040102 kg. On the touch screen, only the basket feeding motor lit up: the transverse conveyor motor and the basket unloading motor did not light up, indicating that the collection basket with the expected capacity had been unloaded to the ground, and the basket feeding motor was transporting empty baskets to the basket unloading conveyor belt. Through bench simulation tests, it was determined that the control system for the automatic unloading and replenishment of baskets based on the cooperative detection control strategy of the photoelectric sensor and pressure sensor reduced the probability of misjudgment and misoperation and improved system performance. This was conducted with the probability of system misjudgment and misoperation serving as the main evaluation index. The simulation results demonstrated that the control system for the automatic unloading and replenishment of baskets based on a photoelectric sensor and pressure sensor cooperative detection control strategy could be error-free judgment and avoid misoperation, effectively improving the stability, accuracy, and rapidity of the system. The study’s findings could suggest a strategy to lessen the workload of operators and increase the operational effectiveness of harvesters for leafy vegetables.

This paper used the 4UM-120D electric leafy vegetable harvester as the research object and designed a travel speed automatic control system to maintain the travel speed within a set value of ±2% in order to improve the efficiency and quality of leafy vegetable harvester operations and decrease the work intensity of the operator. The harvester’s travel speed was automatically controlled by using the PID, adaptive fuzzy PID, and sliding mode control techniques after the mechanical and electrical equations for the travel drive motor (a DC brushless motor) were established in MATLAB. By simulating various working situations, the stability, accuracy, and speed of the automatic control system were compared and analyzed using the adjustment time, overshoot, steady-state transition time, and maximum deviation from the set speed as evaluation indicators. The test results revealed that when the current value of the leafy vegetable harvester travel speed deviated from the set value by more than 2%, the dynamic response performance and stability of the DC brushless motor travel drive system based on the sliding mode control strategy was significantly better than that of the PID and adaptive fuzzy PID control strategies, and its anti-disturbance was stronger, achieving the function of automatic control of the harvester travel speed. When the travel motor started with a constant load and the sliding mode control strategy’s parameters were the gain factors A = 1/70, c = 100, ε = 100, and k = 100, the travel drive system regulation time was 1.5 s, and the overshoot was 10%. When the harvester was operating smoothly and had leafy vegetable collection baskets loaded and unloaded, the steady-state transition time was 0.3 s. According to the actual engineering application experience, the specific technical state of the control strategy of the agricultural machinery travel speed automatic control system was: regulation time 2.5~3 s; overshoot amount 20~25%; and steady-state transition time 1.0~1.5 s, so the travel speed automatic control system of the electric leafy vegetable harvester in sliding mode was in line with the technical state requirements. The results of the field trials demonstrated the accuracy of the simulation test results. This study offered a method to lessen the work intensity of operators and increase the operating efficiency and quality of a leafy vegetable harvester.