-
Review Article
- A review of crop sterilization robots in agriculture: UV sterilization machines
- Donghyeon Kang, Jun-Woo Park, Na-Eun Kim, Ho-Seung Jang, Jeong Gyun Kang, Choung-Keun Lee, Kyung-do Kwon
- In greenhouse crop cultivation, fungal diseases, including powdery mildew, are a major factor in reducing productivity and quality. Existing pesticide control methods, …
- In greenhouse crop cultivation, fungal diseases, including powdery mildew, are a major factor in reducing productivity and quality. Existing pesticide control methods, relying on chemical pesticides, have limitations, including the development of pesticide resistance, pesticide residues, and worker safety. Ultraviolet (UV) sterilization technology, a chemical-free physical pesticide control method, has proven effective in managing diseases in major greenhouse crops such as strawberries, tomatoes, and cucumbers. This paper examines the basic characteristics and sterilization mechanisms of UV and summarizes the technological advancements of UV sterilization equipment used in agriculture. Specifically, through in-depth analysis of field applications conducted overseas, we comprehensively examine the disease suppression effects and plant physiological responses associated with direct UV irradiation. Finally, we propose a phased introduction strategy appropriate for the domestic greenhouse environment, along with an economic feasibility analysis and policy support measures. UV sterilization technology is a key technology for achieving eco-friendly agriculture and enhancing export competitiveness. With appropriate technological development and policy support, it is expected to become the standard pest control technology for domestic greenhouse horticulture by 2030. - COLLAPSE
-
Research Article
- Optimization of design variables for amaranth stem cutting blade using data-driven machine learning model and genetic algorithm
- HoYun Yu, DongGun Lee, SungBo Shim
- This study investigated the optimal geometric design parameters of a cutting blade to minimize stalk cutting energy for improved energy efficiency in …
- This study investigated the optimal geometric design parameters of a cutting blade to minimize stalk cutting energy for improved energy efficiency in amaranth harvesting. As the industrial demand for amaranth grows, mechanized harvesting for mass production has become increasingly necessary, requiring optimized design of the cutting process, which accounts for a significant proportion of total energy consumption. The shear angle, bevel angle, sliding angle, and approach angle were selected as the primary design variables. A dataset of 519 cutting experiments was compiled, comprising 189 response surface methodology (RSM) experiments and 330 one-factor-at-a-time (OFAT) experiments, to analyze variable interactions and the effects of individual factors. The data were preprocessed and used to train a cutting strength prediction model based on the XGboost algorithm, achieving a coefficient of determination (R2 = 0.7098). Variable importance analysis identified the bevel angle as the dominant factor influencing cutting strength. The developed prediction model was then applied as the fitness function of a genetic algorithm to identify the optimal combination of design parameters. The optimization yielded a predicted minimum cutting strength of 0.1787 N/mm2, with a shear angle of 65.59°, a bevel angle of 16.72°, a sliding angle of 11.45°, and an approach angle of 37.78°. These results validate the proposed methodology combining machine learning and genetic algorithm optimization as an effective strategy for deriving cutting blade design parameters beyond the limitations of physical experimentation alone. - COLLAPSE
-
Research Article
- Development of a post-harvest conveying and loading module for apple harvesting robots
- Do Yeon Im, Uk Tae Choi, Dae Hee Kim
- Automation of apple harvesting in open-field orchards has attracted attention as a key technology for alleviating rural labor shortages and improving harvesting …
- Automation of apple harvesting in open-field orchards has attracted attention as a key technology for alleviating rural labor shortages and improving harvesting efficiency. However, apples are easily damaged by impact, free fall, and localized pressure during post-harvest conveying and loading processes. Therefore, the performance of an apple harvesting robot is strongly affected not only by fruit recognition and gripping technologies but also by the stability of the post-harvest handling system. In this study, a post-harvest conveying and loading module applicable to an apple harvesting robot was developed, and an integrated control system for operating the module was implemented. The developed module consists of an infrared sensor-based fruit receiving unit, a conveyor-based conveying unit, and an iris mechanism-based loading unit. To reduce fruit impact during conveying, cushioning material was applied to the contact surface of the conveyor. During loading, the loading unit temporarily supports the fruit, moves it to the target position, and discharges it into the collection box, thereby reducing the free-fall height and localized accumulation of fruits. The developed system provides a basis for future quantitative evaluation of conveying stability, loading uniformity, and fruit damage reduction through integration with an apple harvesting robot platform. - COLLAPSE
-
Research Article
- Analysis of work efficiency and economic feasibility of a potato collector and follow-type transporter through field performance test
- JoonYong Kim
- The collecting operation during the potato cultivation process in Korea is one of the most manpower-dependent operations. With the intensifying issues of …
- The collecting operation during the potato cultivation process in Korea is one of the most manpower-dependent operations. With the intensifying issues of an aging rural population and labor shortages, mechanization has become an urgent necessity. In this study, the work efficiency of a developed potato collector and a follow-type transporter was measured, and its economic feasibility was analyzed by comparing it with conventional manual operation. Field test was conducted at experimental field in Gangwon-do, Jeongseon-gun to measure the operating times of mechanized and manual collecting operation. The work efficiency evaluation revealed that mechanized operation required 0.97 h/10a, demonstrating an approximately 3.7 times improvement compared to manual operation (3.59 h/10a). For the economic analysis, the annual fixed and variable costs were calculated using the machine cost analysis method. The total annual cost for the mechanized operation was 12,825,673 KRW/yr, whereas the manual operation cost was 18,219,360 KRW/yr, indicating that mechanization reduces annual cost by 5,393,687 KRW. Furthermore, a break-even point analysis based on cumulative cost showed that the mechanized system secures an economic advantage over manual labor starting in 9.3 years after its introduction. - COLLAPSE
-
Research Article
- Development of DEM models for garlic field soil and whole garlic bulbs based on physical and contact properties
- Jae-Seo Hwang, Young-Woo Do, Si-Eon Lee, Yi-Seo Min, So-Yun Gong, Seon-Ju Park, Gi-Deok Kim, Jin-Woo Park, Hyeon-Seo Yoon, Wan-Soo Kim
- Mechanization of garlic harvesting is required to address labor shortages and rising production costs. However, the design of garlic harvesting machinery remains …
- Mechanization of garlic harvesting is required to address labor shortages and rising production costs. However, the design of garlic harvesting machinery remains challenging because digging performance is governed by complex interactions among soil, garlic bulbs, and machine components. The discrete element method (DEM) can be used to analyze these interactions, but reliable simulations require calibrated material and contact parameters that can reproduce measured macroscopic responses. In this study, DEM models of soil and garlic were developed and calibrated based on measured physical and contact properties. Soils and garlic bulbs were collected from Hapcheon (southern-type) and Uiseong (northern-type), two major garlic-producing regions in Korea. Soil properties, including texture, bulk density, and shear behavior, were measured, while three-dimensional garlic bulb models were constructed using photogrammetry. The physical and contact properties of garlic were also measured. The soil DEM models were calibrated by reproducing the measured bulk density and vane shear torque, whereas the garlic DEM models were calibrated by reproducing the measured critical angle and rebound height. The two fields showed distinct differences in soil texture (silt loam and clay loam), bulk density (1,266.1 and 1,360.4 kg/m3), and mean vane shear torque (8.17 and 20.45 N·m). After calibration, the bulk density and vane shear torque errors were reduced to within 0.13% and 0.93%, respectively, while the critical angle and rebound height of garlic were reproduced with errors of less than 5%. Finally, the calibrated soil and garlic models were integrated into a DEM simulation environment. The developed models provide fundamental data and calibrated DEM inputs for future analysis of soil–garlic interactions. - COLLAPSE
-
Research Article
- Optimization of alfalfa cutting parameters using response surface methodology and kriging model
- In-Beom Son, Dong-Gun Lee, Sung-Bo Shim
- This study analyzed the effects of blade geometry and cutting speed on the cutting force during the alfalfa stem cutting process and …
- This study analyzed the effects of blade geometry and cutting speed on the cutting force during the alfalfa stem cutting process and derived the optimal combination of blade parameters to minimize the cutting force. To this end, a four-factor, three-level fractional factorial design was conducted using shear angle, approach angle, sliding angle, and blade speed as factors to acquire force-time signals. The maximum cutting force was calculated by applying a Savitzky–Golay filter to the acquired signals, and a quadratic response surface model was constructed using this as the dependent variable. As a result, the coefficient of determination (R2) of the model was 0.8489. Analysis of variance (ANOVA) results showed that the overall model, the main effect and quadratic term of the approach angle, and the interaction term between the shear angle and speed were statistically significant. In the response surface analysis by speed, the minimum cutting force region was formed around an approach angle of 40° under all conditions of 100, 300, and 500 mm/min. As the speed increased, the optimal shear angle tended to shift from 60° to 40°, and the optimal sliding angle shifted from 0° to 45°. Furthermore, a Kriging model was employed to capture the non-linear characteristics of the cutting process. Through non-linear optimization based on the Kriging model, global optimal conditions minimizing the cutting force within the design space were precisely derived, thereby providing quantitative criteria for designing alfalfa cutting blades that combine speed, shear angle, and sliding angle based on an approach angle of 40°. - COLLAPSE
-
Research Article
- A virtualized DUT-in-the-loop test harness and a differential conformance-judgment methodology for standards certification of greenhouse integrated controllers
- JoonYong Kim
- A greenhouse integrated controller is the gateway device that mediates, under the Korean Standards KS B 7958/KS X 3267/3288, between farm management …
- A greenhouse integrated controller is the gateway device that mediates, under the Korean Standards KS B 7958/KS X 3267/3288, between farm management software and the field sensors, actuators, and sub-nodes. Certifying that such a controller conforms to these standards traditionally requires a physical test bed of real devices and RS485 buses, which is costly to assemble, hard to reproduce between test rounds, and ill-suited to injecting the faults a certification body must exercise. A further difficulty is that the configuration a controller reports upward need not match the devices actually installed in the field, and this mismatch is itself a subject of the test. We present a virtualized device-under-test (DUT)-in-the-loop test harness that keeps only the controller as real hardware and reproduces the entire field topology on a commodity computer. The harness splits into two agents: a Northbound agent that observes and commands the DUT over Modbus exactly as farm management software would, and a Southbound agent that simulates the field devices and therefore holds the ground truth. Building on this separation, we define a differential conformance-judgment methodology, where “differential” means that each verdict compares the DUT’s report against the simulated field truth, rather than checking the DUT’s outputs only for internal self-consistency. Control authority over a device is observed on up to six axes (three reported by the gateway, three from nodes as ground truth), and verdicts are rendered for command acceptance, mirroring accuracy, and sensor/actuator readings, with traceability from each verdict down to the captured Modbus frames. Evaluated on three commercial controllers across gateway and flat multi-node topologies, the methodology surfaces defects that internal self-consistency cannot resolve: it distinguishes a genuine defect from a standard-conformant rejection, and exposes a controller that reports a command as applied while its view of the field has gone stale. - COLLAPSE
-
Research Article
- Comparison of equal- and CRITIC-weighted desirability functions for multi-response optimization of blade angles in amaranth stem cutting
- SungHun Jung, DongGun Lee, SungBo Shim
- The production and commercialization of amaranth have expanded due to increasing interest in its nutritional value and environmental adaptability, creating a need …
- The production and commercialization of amaranth have expanded due to increasing interest in its nutritional value and environmental adaptability, creating a need to improve the cutting performance of harvesting machinery. Improving cutting performance is closely related to determining the optimal blade conditions, and multi-response optimization considering both force and energy indices is required. In multi-response optimization, the desirability function is commonly applied using equal weights. However, when responses are highly correlated or when the relative importance of each response is difficult to determine in advance, the use of equal weights may lead to biased optimal conditions. Therefore, in this study, the model-predicted optimal cutting conditions for amaranth stems were derived using the Criteria Importance Through Intercriteria Correlation (CRITIC) weighting method and compared with those obtained using the equal weighting method. As a result, both methods produced the same model-predicted optimal conditions: a shear angle of 40.00°, an approach angle of 40.00°, and a bevel angle of 10.00°. Under these conditions, the predicted cutting strength and specific cutting energy were 0.06798 N/mm2 and 0.41062 mJ/mm2, respectively. The overall desirability was 0.90797 for the equal weighting method and 0.90858 for the CRITIC weighting method, showing a slightly higher value when the CRITIC weighting method was applied. However, the identical optimal conditions obtained by the two methods were attributed to the fact that both responses decreased toward the same boundary of the design space and that the difference in the normalized standard deviations of the two responses was small. This study demonstrated the feasibility of the CRITIC weighting method for optimizing the cutting conditions of amaranth stems, and further multi-response optimization studies including additional responses such as blade wear and cut-surface quality are needed. - COLLAPSE
Journal Informaiton
Precision Agriculture Science and Technology
Precision Agriculture Science and Technology
Korean Society of Precision Agriculture
한국정밀농업학회
대구광역시 북구 대학로 80, 경북대학교 농업생명과학대학 4호관 605호
Tel: 053-950-6891 / E-mail: kspaknu21@gmail.com Copyright© Korean Society of Precision Agriculture. Powered by APUB
Tel: 053-950-6891 / E-mail: kspaknu21@gmail.com Copyright© Korean Society of Precision Agriculture. Powered by APUB