Due to the cutting mechanism of the existing grafting machine, it cannot adjust the cutting angle in real time, resulting in low fitting precision on the cutting surfaces between the rootstocks and scion seedlings and, thus, seriously affecting the survival rate and quality of the grafting seedlings. In this paper, a kind of splice grafting method based on visual image is proposed, aiming at maximizing the joint rate between cutting surfaces of rootstocks and scion seedlings and realizing precise cutting and grafting of grafting machine. After analysis, we determined that melon rootstock seedlings have a structure of pith cavity inside, and the solid structure from the top of the pith cavity to the left and right base points of a growing point forms the important area of a cutting surface. In order to obtain the geometric model of the cutting surfaces of the seedlings, a visual image analysis system was established to identify, analyze, and model the pith cavity structure inside the rootstock seedling, as well as the external morphological characteristics, and the ultimate cutting angle of the rootstock seedling and cutting surface parameters were determined. By measuring the length of minor axis of scion seedlings in order to achieve the maximum joint rate, the optimal cutting angle of the rootstocks and scion seedlings was determined. Then grafting and seedling cultivation tests were carried out. The test results showed that the range of ultimate cutting angle on rootstock seedlings (Cucurbita moschata) was 18.21 ± 1.92°; the cutting angles of the rootstock (Cucurbita moschata) and scion seedlings (watermelon) were 22° and 19.68°, respectively; the cutting surface length of the two was 4.96 mm; and the cutting surface thickness of the rootstock was 0.13 mm, all of which could satisfy the technological requirements of the matched splice grafting of melons. The research results can serve as a reference for the design in vision-guided precision cutting and real-time grafting operation on grafting robots.

Design and optimization on rootstock cutting mechanism of grafting robot for cucurbit

In view of the need to remove empty cells and unqualified seedlings for automatic transplanting of leafy vegetable seedlings, this paper proposes a method to detect the growth parameters of leafy vegetable seedlings by using machine vision technology. This method uses the image processor PV200 to perform image grayscale, threshold segmentation, corrosion, expansion, area division, etc. to obtain the pixel value of the leaf area of the seedling and compare it with the set standard value, which provides guiding information for eliminating empty cells and unqualified seedlings. Lettuce seedlings at 17 days, 20 days, and 22 days of seedling age were used as the test objects, and the growth status and test results of the seedlings were analyzed to determine the optimum seedling age for transplanting. The test results show that there is basically no leaf cross-border between the lettuce seedlings at the age of 17 days, the average pixel area of the leaves is 3771.74, and the detection accuracy rate is 100%; the seedlings at the age of 22 days grow 5–6 leaves, the detection accuracy of unqualified seedlings and qualified seedlings was 62.50% and 88.16%, respectively, and the comprehensive detection accuracy was 85.71%. The comprehensive detection accuracy rate showed a downward trend with the increase of seedling age, mainly due to the partial occlusion between leaves. The transplanting of leafy vegetable seedlings is a sparse transplanting operation, and the seedling spacing increases after transplanting. Therefore, the detection of seedlings in the process of transplanting can greatly improve the recognition accuracy and solve the problem that the leaves of the seedlings in the seedling tray are obscured by each other and affect the detection accuracy. The research results can provide a theoretical basis and design reference for the development of the visual inspection system and the transplanting actuator of the leafy vegetable seedlings transplanting robot.

The rising age of the population in rural China and the labor intensity of grafting have resulted in a decrease in the number of grafters and a subsequent increase in their wages. Manual grafting can no longer satisfy the increasing demand for watermelon-grafted transplanting; thus, machine grafting will be an effective alternative. In order to accelerate the implementation of machine grafting in China, a comparative analysis between the automatic grafting machine (model 2TJGQ-800) and traditional hand grafting was conducted. The reliability and feasibility of machine grafting were evaluated through a comprehensive evaluation of the production capacity and grafting seedling quality. This study focuses on the grafting application of watermelon plug-tray seedlings. The scion and rootstock seeds were sown on 9 November 2022. Grafting experiments using an automatic grafting machine, skilled workers, and ordinary workers were conducted with the root-pruned one-cotyledon grafting method on 24 November 2022. The results showed that the machine grafting had a high uniformity and grafting speed. The grafting speed of the grafting machine was 774 plant·h⁻¹ and 1.65–2.55-fold higher than the hand grafting. With training, workers can improve their grafting speed, but it will still be slower than machine grafting. In addition, there was no significant difference in the grafting survival rate between the machine grafting and hand grafting. However, using machine grafting, the success rate decreased from 100% to 90.07% and the rootstock regrowth rate increased from 18.44% to 72.69%. Incomplete rootstock cutting, clip supply failure, and grafting drop failure are the three main factors that result in machine grafting failure. In conclusion, the grafting machine has advantages in terms of grafting speed and uniformity. Upon improving the accuracy of the cutting mechanism and grafting success rate, it will be adopted by commercial nurseries.

Aiming to solve the problems of poor performance and low stability in the automatic clip-feeding of a grafting machine, an automatic clip-feeding mechanism with a precise single-clip discharge mechanism was designed, and a clip-feeding performance test was carried out. Taking the grafting clip of the 2TJGQ-800 type of vegetable-grafting robot as the research object, the clamping-force analysis model of the grafting clip was constructed by ABUQUS finite-element analysis software, and the variation law of clamping force, steel wire diameter, and opening deformation, as well as the calculation equation of clamping force, were obtained. The grafting clip model was verified by mechanical test, and test results showed that the grafting clip with a steel wire diameter of 0.7 mm proved safe and reliable for grafted cucumber and watermelon seedlings; the grafting clip with steel wire diameter of 0.8 mm had a risk of producing injury to grafted cucumber and watermelon seedlings when clamping. The method of single-clip discharge in the inclined discharging slideway was put forward, and the components for clip discharge and clip pushing were designed. The critical thrust for sending out the grafting clip in the clip-feeding slideway was 0.603 N after analyzing the force status of the grafting clip in the clip-feeding slideway. Test results showed that the success rate of automatic clip-feeding reached 98.67% when inclination angle of row-discharging slideway was 50° and the thrust of clip-pushing cylinder (input air pressure of 0.4 MPa) was 8.04 N, which met the technical requirements of mechanical grafting. The inclination of the grafting clip and the damaged clip in the feeding slideway is the main reason for the failure of clip-feeding. The research results can provide theoretical and design references for the innovative research of the automatic clip-feeding mechanism of grafting robots.

The grafting machine cutting step is the core step of the grafting process. The existing grafting machine cutting mechanism adopts fixed angle cutting and manually adjusts the cutting angle based on experience, and the cutting angle is not definite for rootstock cutting in different growth periods. In this paper, we proposed a method to guide the precise cutting mechanism according to internal and external parameters of melon rootstock at a specific period. First, we constructed a cutting model based on internal and external characteristics of rootstock cutting in the growth period of “two leaves and one core” and clarified the safe cutting area. Second, we designed a rotary precision cutting mechanism for rootstock cutting with automatic angle adjustment and constructed the displacement equation of the cutting trajectory of the cutter according to the cutting model. Last, we examined the cutting effect of the precision cutting mechanism and determined the precise cutting angle of the rootstock cutting mechanism in the growth period. Finally, the cutting effect of the precision cutting mechanism was examined, and the precision cutting angle of the rootstock during the growth period was determined. A comparison test between the precision cutting mechanism and the traditional cutting mechanism was carried out, and visual images of the incision were captured and analyzed. The results show that under the five horizontal cutting angles in the safe cutting area of rootstock, the length of the cut surface is inversely proportional to the cutting angle, and the flatness of the cut surface is directly proportional to the cutting angle. Comprehensive evaluation of the length of the cut surface, the cutting success rate, and the quality of cutting revealed that the average cutting angle of the precision cutting mechanism in the safe cutting area (26°) is better than that of the traditional cutting mechanism. It meets the technical requirements of the cutting technology for mechanically grafted rootstocks. The results provide a reference for studying new rootstock precision cutting mechanisms and cutting angle adaptive control models.