The overall 3D structure of the 12-kW self-propelled riding-type automatic onion transplanter investigated in the study, which comprises two key components: an operator-driven four-wheel vehicle and a transplanter, is shown in Fig. 1a. The onion transplanting steps include three primary mechanisms, the same as other vegetable transplanters, which are seedling picking, rotary seedling conveying and planting (Chowdhury et al., 2021; Reza et al., 2021). The six-row onion transplanter is composed of six picking assembly units, conveyors, and a rotary planting mechanism, as shown in Fig. 1b. The picking mechanism (Fig. 1b-1) combines the mechanism of pushing pins and the seedling conveying mechanism. The conveyor system (Fig. 1b-2) delivers seedlings to the rotating planting hopper. Three primary linkages and a planting hopper compose the rotating planting mechanism (Fig. 1b-3). The rotary planter requires a constant supply of onion seedlings at a certain speed to ensure regular planting. Each rotating hopper assembly transplants one seedling during a single transplantation cycle. There are two wheels for each row of seedlings, so the soil is finally pressed down to make sure there is no mulch film damage and that the seedlings are standing upright. Six seedlings are taken from the tray throughout a complete cycle of the pushing bars unit assembly and are then conveyed to the seedling conveyor via the seedling carrying assembly.

The seedlings are pushed into the hopper by the pushing bar lever, which operates in tandem with the seedling conveyor to provide continuous motion between the conveyor and the pushing bar. Due to the constant motion of the conveyor, it is necessary for the planting mechanism to move in a continuous manner in both the upward and downward directions. Pushing and carrying assemble, on the other hand, experience motion pauses during the vertical motion of the tray and at the picking position of the seedling, respectively.

An elliptical motion trajectory is produced by the proposed vertical and horizontal rotary planting mechanisms. When the hopper reaches its top position during the motion, it receives the falling onion seedlings from the conveyor. At the bottom of the rotation cycle, the hopper punches the seedlings into the soil. Fig. 2 illustrates the schematic view of the seedling collection from the conveyor and their subsequent transplantation into the soil. The rotational speed of the planting mechanism and the forward speed of the transplanter are the key factors in determining the optimal working speed for the transplanter.

Rotating speed of the planting mechanism

In order to design the planting mechanism, it is vital to know the rotational and forward speeds. The rotational speed of the planting mechanism depends upon the forward speed of the transplanter and the required plant-to-plant distance for a specific crop. The proposed transplanter consisted of six rows with a spacing of 0.15 m between them, and the required planting interval was 0.13 to 0.14 m. A planting mulch bed of 0.09 m width could be considered for the transplantation of onion seedlings (Reza et al., 2021). Table 1 shows notations and definitions with the units of the variables used in this study.

The rotating speed (n, rpm) of the planting mechanism can be calculated using Eq. (1), which is a function of the planting interval (P_i, m) and the forward speed of the transplanter (v_t, m s^-1):

Srivastava et al. (2006) suggested Eq. (2) to determine the planting supply rate (R_st, seedlings min^-1) of the mechanism:

Another crucial criterion for the transplanting operation is the insertion of seedlings into the hopper, with the straight and vertical orientation of the seedlings being vital. Successful feeding of the onion seedlings into the hopper is dependent on the free-falling distance, the seedling supply rate, and the amount of time necessary to fall. During the planting procedure, the most crucial factor to consider is the synchronization of all of these parameters with the rotating speed of the hopper. The deposition should take place at the highest vertical position in order to reduce the amount of seedling fall time. The required falling time (F_t, s) of the onion seedlings can be found using Eq. (3):

The free-falling velocity (v_s, m s^-1) of the seedlings is limited by the aerodynamic drag and is determined by the physical characteristics of the seedlings, as in Eqs. (4) and (5) (Manilla & Shaw, 1987):

where α is a constant which can be calculated as in Eq. (5):

This study used 50-day-old onion seedlings to analyze the physical properties of the onion seedlings because it has been proven that this age is most suitable for mechanical transplanting (Kim et al., 2015; Dihingia et al., 2016).

Constant horizontal speed of the planting mechanism

Constant horizontal velocity, also called zero velocity, occurs at the maximum lower position of the planting hopper, where the seedling undergoes the deposition process into the soil. The hopper velocity should match the forward speed of the planter in order to release the seedling at the desired depth (Manilla & Shaw, 1987). Due to the constant horizontal velocity, there is no acceleration at the lowest point of the hopper, which enables the hopper to deposit seedlings at the desired depth into the soil surface. As a result, constant horizontal velocity should be equal but opposite to the onion transplanter’s forward speed. Eq. (6) shows the relationship between the forward velocity of the transplanter (v_t, m s^-1) and the horizontal velocity of the hopper (v_x, m s^-1):

During the transplanting operation, the working speed trajectory of the mechanism may provide a more precise outline for measuring the zero-speed delivery of the seedlings. To develop the working speed trajectory of the planting mechanism (Fig. 3), a mathematical model was derived using a Cartesian coordinate system. The forward speed of the onion transplanter was considered the negative direction of the horizontal x-axis and the positive direction along the vertical y-axis, and the lowest point of the hopper was considered the origin point. The mathematical model can be expressed using Eqs. (7) and (8):

where R_t is the rotational radius. The rotational radius can be calculated using Eq. (9) of the rotary planting mechanism:

In order to minimize the damaging effects of the mulching film on the planting bed, the characteristic coefficient factor (λ) was considered, which is the ratio between the circumferential speed of the planting mechanism and the forward speed of the transplanter (Du et al., 2018), expressed using Eq. (10):

The plastic mulch damage is minimal when the above-mentioned ratio is equal to 1. In this scenario, the circumferential rotational speed should be the same as the forward rotational speed. As a result, it is essential to choose a working speed that is compatible with the circumferential speed while also satisfying the required characteristic coefficient factor requirement.

Simulation of motion and working speed

To determine the working speed of the proposed planting mechanism, a design and motion study were conducted using a commercial simulation software package (SOLIDWORKS 2018, Dassault Systems SolidWorks Corp., Waltham, MA, USA). A virtual 3D prototype model was developed, as shown in Fig. 4, and tested at different forward speeds (0.10, 0.15, and 0.20 m s^-1) within the optimum human working comfort range. The rotating speed of the mechanism and the forward speed of the transplanter were the two most important simulation factors. Trajectories of several forward speeds were recorded and analyzed in order to determine the forward speed that resulted in the least amount of mulch film damage during the operation. The rotating speed of the transplanter mechanism and the forward speed of the transplanter were found to be 40 to 80 rpm, and 0.10 to 0.20 m s^-1, respectively. The optimal working speed of the planting mechanism was determined.

Validation with a prototype

The field tests were done at the Rural Development Administration (RDA) in Jeonju, Korea (lat. 35.84°N, long. 127.13°E). To validate the simulation findings, a prototype of the gear-driven rotating planting mechanism was fabricated and operated on a rotational soil bed test bench (Fig. 5). To ensure effective seedling deposition in the hopper, 50-day-old onion seedlings with an average height of 237.3 ± 12.4 mm and weight of 11.23 ± 1.5 g (plant and soil) were used.

The diameter, width, and depth of the rotating soil bed were 6.5 m, 0.45 m, and 0.2 m, respectively. The soil bed rotated at a speed of 0.3 to 1.8 m s^-1, and the planting mechanism was driven by a 120 W 9BDG5 electromagnetic braking motor (DKM Motors Co. Ltd., Incheon, Korea). A SV-iG5A inverter (LS Industrial Systems Co., Anyang, Korea) was used to regulate the speed of the electric motor.

The soil properties on the test bench were tested with five replications. The soil sample on the test bench was sandy loam, comprising 78 ± 8% sand, 14 ± 2% silt, and 8 ± 6% clay. Soil temperature, water content, electrical conductivity, and cone index were measured. A handheld cone penetrometer Penetrologger 6.00 (Eijkelkamp, Giesbeek, The Netherlands) was used to measure the Cone index up to 0.15 m with the error less than 1%. The soil temperature, water content, and soil electrical conductivity (EC) were measured with an error less than ± 0.5oC, ± 1%, and ± 0.1 dS m^-1, respectively, using a WT 1000N soil sensor (RF sensor, Seoul, Korea). The average bulk density, cone index, soil water content, soil temperature and soil EC were found to be 1.31 ± 0.08 g cm^-3, 0.58 ± 0.07 MPa, 23.17 ± 0.97%, 29.7 ± 0.45oC, and 1.31 ± 0.15 dS m^-1, respectively.

To determine the power requirement of the planting operation, a TRS605 torque sensor (FUTEK Co., Irvine, CA, USA) and a SEN041F tri-axial acceleration sensor (PCB Piezotronics, Inc., Depew, NY, USA) were installed in the power transmission line, between the motor and the planting mechanism, and in the planting hopper jaws, respectively. A NI 6212 data acquisition device (National Instruments, Austin, TX, USA) was used to acquire the signals from the sensor. We used a LabVIEW 2018 software tool (National Instruments, Austin, TX, USA) to collect torque and acceleration data. The data obtained from the torque sensor underwent a smoothing process using a symmetric moving average method (Rohrer et al., 2018) with a window size of 20 points. Another software package MATLAB R2019a (The MathWorks, Natick, MA, USA) was used to analyse the collected data. All sensors were mounted on the planting mechanism of the test bench, as illustrated in Fig. 5b.

To assess the effectiveness of the planting mechanism, onion seedlings were planted at forward speeds of 0.10, 0.15, and 0.20 m s^-1, with a rotational speed of 60 rpm. The study focused on measuring two key parameters: the misplanting rate and the uprightness of plant. The misplanting rate refers to instances where the seedlings were not positioned correctly. Plant uprightness was determined by examining the angle between the stem and the ground, considering angles below 30ᵒ as indicating proper upright positioning by Han et al. (2019). The transplanting operation was performed five times on the test bench, as depicted in Fig. 5. During the experiment, the misplanting rate and planting interval uniformity were evaluated. Plant intervals were measured using a measuring tape, while the angle between the stem and ground (plant uprightness) was measured using a protractor. To ensure accuracy, all measurements were taken with five replications. Descriptive statistical analysis methods (average, one-way ANOVA, least significant difference (LSD) all-pairwise comparisons test) were used in this study. A software package Statistix 10 (Analytical Software, Tallahassee, FL, USA) were used for the statistical analysis.

The rotational speed of the planting mechanism in the onion transplanter is influenced by the forward speed of the machine and the desired planting interval between onion seedlings. In this study, considering a range of forward speeds from 0.10 to 0.25 m s^-1 and a required planting interval of 0.14 m for the onion seedlings, the calculated range of rotational speeds for the planting mechanism was found to be between 50 and 70 rpm as shown in Fig. 6a.

To ensure proper seedling drop into the hopper, the height at which the seedlings fall is a critical factor. Higher falling heights result in increased air drag on the seedlings, potentially causing improper seedling placement. Additionally, the clearance between the seedling release point and the hopper’s operating route needs to be considered. Considering these factors, a free-falling height of 0.08 m was determined. The planting hopper speed was then calculated to be 60 rpm, as illustrated in Fig. 6b.

Validation tests conducted at a free-falling height of 0.08 m demonstrated a high success rate of 97.16% for seedling deposition into the hopper. The experiment revealed that seedlings with a straight shape and fewer leaves exhibited a higher percentage of accurate placement. It was also observed that the vibration of the transplanter during the operation played a significant role in seedling displacement during free fall. These findings provide valuable insights for optimizing the rotational speed of the planting mechanism, ensuring proper seedling placement and minimizing errors during the onion transplanting process.

Fig. 7 presents the simulated trajectories of the planting mechanism for the transplanter forward speeds of 0.10, 0.15, and 0.20 m s^-1, at 60 rpm of the rotational speed of the planting hopper. The analysis revealed that the optimal performance, characterized by smooth and accurate seedling deposition, was achieved at a forward speed of 0.15 m s^-1 which provide most favourable outcome in terms of successful seedling placement.

Three different scenarios were investigated for the value characteristics coefficient, namely, λ>1, λ=1, and 0<λ<1 for transplanter forward speeds of 0.10, 0.15, and 0.20 m s^-1, respectively. The planting mechanism trajectories for these scenarios are depicted in Figs. 7a, 7b, and 7c.

In the first scenario, when the forward speed of the transplanter was 0.1 m s^-1, the characteristic coefficient was higher than 1. The trajectory of the planting mechanism was affected by the void deposition of the onion seedlings. There was an inflection point at the lowest position of the cycloid created under the soil surface. The direction of the horizontal speed on the cycloid was the same as the direction of the transplanter’s forward speed, with the exception of the point of inflection. The speed at the point of inflection can meet the condition of the constant horizontal speed principle. However, the seedlings were rammed forward because the point of inflection was located at the lowest position of the trajectory, and the cycloid immediately moved forward after delivering the seedlings at the point of inflection. Because of the height of the seedlings, it took time for them to pass away from the cycloid. The seedlings with the cycloid progressively going ahead had not entirely gone away from the cycloid, causing damage to the seedlings during the operation. Additionally, the planting mechanism created a hole for depositing the seedlings within the soil. It entered at one location and exited at another, resulting in unwanted mulch film damage. As shown in Fig. 7a, the condition of λ>1 was not optimal for delivering the onion seedlings.

With the scenario of λ=1, when the transplanter’s forward speed was 0.15 m s^-1, the trajectory beneath the soil became a straight line, indicating that the hopper entered and exited at the same point in the soil after releasing the seedlings. This is because the speed of the planting hopper at the seedling deposition point was equal to and opposite the forward speed of the transplanter. The seedlings were deposited at a constant horizontal speed, as shown in Fig. 7b, and the seedlings were able to maintain the appropriate degree of uprightness throughout the depositing process. Since the planting mechanism entered and departed from the same point, there was a little damage to the mulch coating. The characteristic coefficient λ=1 was an absolute requirement for the onion transplanter to function properly, which was assumed to move at a forward speed of 0.15 m s^-1.

The direction of the movement of the seedling could be indicated by the tangent point on the working trajectory. In the scenario of 0<λ<1, where the forward speed of the planter was 0.20 m s^-1, the trajectory of the hopper tip beneath the soil surface created a short cycloid as shown in Fig. 7c. It was observed that the direction of horizontal speed at each location on the short cycloid was identical to the direction of forward speed of the transplanter. This shows that the horizontal speed and the angle of the onion seedling deposition were critical, as there was little time to deposit and drag out of the soil. Thus, constant horizontal speed delivery is impossible in this case, and different impact points caused damage to the mulch coating by the transplanter.

Furthermore, the cycloid movement has the potential to retain the seedlings in an upright position. In the field test, it was found that this circumstance resulted in the highest straightness of the transplanted seedlings, as shown in Fig. 8. Moreover, the cycloid movement could keep the seedlings in an upright orientation. In the field test, maximum uprightness of the planted seedlings was also found in this condition (Fig. 8). As the planting mechanism’s entry and exit points were the same, the mulch film’s damage was minimal. Thus, having the characteristic coefficient equal to 1 was a necessary prerequisite for performing typical transplantation procedures with the machine. According to these findings, for the onion seedlings to be planted straight and perpendicular, it is necessary to maintain a 1:1 ratio between the rotational speed of the planting mechanism and the forward speed of the transplanter throughout the operation. During the transplanting work in the field, it would also be helpful to minimize the physical labor and seedling damage.

Plant-to-plant distance for onion seedling transplantation depends upon the rotational speed (rpm) of the rotary mechanism and the forward speed of the transplanting machine. Plant interval was measured for different combinations of rotational speed (rpm) of the planting mechanism, and the forward speed of the transplanter is shown in Table 2. Fig. 9 shows the results of the several planting intervals measured during the field experiments.

The usual plant-to-plant distance proposed for onion transplantation was 0.13-0.14 m. Therefore, the planting interval obtained with the combination of 60 rpm rotational speed and 0.15 m s^-1 of forward speed of the planting mechanism is 0.13 m, which is suitable for successful transplantation of the onion seedlings. A transplanting rate of 60 seedlings min^-1 can be obtained using above-mentioned combination.

During the transplantation process, the performance of the transplanter was evaluated at three distinct forward speeds (i.e., 0.10, 0.15, and 0.20 m s^-1), and the necessary input power was measured at each of the three different forward speeds. The measured power data for the planting mechanism exhibited a constant reciprocating curve throughout the transplanting process, as shown in Fig. 10.

For the three different forward speeds, the power consumptions of the dibbling mechanism were recorded as 43.07 ± 1.23 W, 36.53 ± 1.34 W, and 42.72 ± 2.29 W, respectively. The power consumption was found to be low when the transplanter was operating at a forward speed of 0.15 m s^-1. The horizontal speed of the planting hopper was developed to the dibbling point where it was able to counterbalance the forward speed. As a result, there was no external force acting on the planting hopper throughout the seedling delivery process due to the forward speed. For the operating forward speeds of 0.10 and 0.20 m s^-1, the planting hopper was pulled in both the backward and forward orientations, respectively. Soil contact with the planting hopper occurred during the seedling’s deposition into the soil, and the tension on the planting hopper changed according to the forward speed conditions. When the forwards speeds were 0.10 m s^-1, the planting hopper created a cycloid inside the soil, forcing it to dig a larger hole with a relatively large area, costing more power. With a forward speed of 0.20 m s^-1, a larger whole area was also dug, but as it towed forward, a part of the forward speed assisted in lowering the power usage. This explained the cause of the power consumption of the planting mechanism was somewhat lower at 0.20 m s^-1 than at 0.10 m s^-1 forward speed.

The planting mechanism completed one cycle of planting in less than one second. Therefore, it shows the minimum and maximum power values were found at the picking and planting positions, respectively (Fig. 10). At the picking and planting positions, the plating mechanism was held in the air and the hopper struck the soil, resulting in a minimum and maximum power consumption, respectively.

Field test results with the forward speeds of 0.10, 0.15, and 0.20 m s^-1 and a rotational speed of 60 rpm of the planting mechanism are shown in Table 3. Table 3 presents the results of a statistical analysis aimed at assessing the impact of forward speed on various parameters of the transplanter, including seedling upright angle, misplanted rate, and maximum power consumption. The findings indicate that the forward speed significantly affects the seedling upright angle, implying that the angle at which the seedlings are positioned was greatly influenced by the speed at which the transplanter moves forward.

Furthermore, the analysis reveals that both power consumption and misplanted rate did not exhibit significant variations in relation to the forward speed. This suggests that the rate at which seedlings are improperly placed and the maximum power consumed by the transplanter remained relatively stable, regardless of the speed at which it operates. These findings highlight the importance of selecting an appropriate forward speed to ensure optimal seedling positioning, while also indicating that power consumption and misplanted rate may be less influenced by variations in forward speed.

According to the findings of the comparison between theoretical and experimental data, the forward speed of 0.15 m s^-1 provided the optimum performance. The upright seedling angle was reported to be 90o ± 2.79o, at a forward speed of 0.15 m s^-1, with a success deposition rate of 92.34%. Different studies carried out transplanting operations with several planting mechanisms at 0.15 to 0.25 m s^-1 forward speed and observed the success rate of transplanting at 90-94% (Tian et al., 2010; Jin et al., 2020; Zhou et al., 2020; Iqbal et al., 2021). To ensure proper placement of seedlings, a free-falling height of 0.08 m was determined. Various studies (Dihingia et al., 2016; Jo et al., 2018) demonstrated that the recommended range for the free falling height of a transplanter is between 0.06 to 0.09 m. The results of the success rate of transplanting in our study were comparable to those of earlier research, despite the fact that the transplanter’s forward speed and the seedling types were different. The results of the power consumption experiment revealed that when the characteristic coefficient λ=1, the planting mechanism used the least amount of power.