Grafting is an effective way to overcome the obstacles of continuous soil cropping and improve the tolerance of plants to abiotic and biotic stresses. An automatic grafting robot can effectively improve the grafting efficiency and survival rate of grafted seedlings, which is an important demand for the commercialization and promotion of vegetable planting. Based on the six main grafting technologies, this paper deeply summarized and analyzed the research status, technical characteristics, and development trends of vegetable grafting robots developed by various countries in the world. At the same time, it focused on the design methods and characteristics of key components such as seedling picking device, clamping device, and cutting device of vegetable grafting robots in detail. Then, the application of machine vision in the grafting robot was compared from the aspects of seed information feature recognition, automatic seedling classification, seedling state detection, and auxiliary grafting. It also was pointed out that machine vision technology was the only way to realize the fully automated grafting of vegetable grafting robots. Finally, several constraints, such as the limited grafting speed of vegetable grafting robots were pointed out, and the future development direction of grafting robots was predicted. As a result, it is believed that the intelligence degree of vegetable grafting robots needs to be improved, and its research and development fail to integrate with the seedling biotechnology, which leads to its poor universality. In the future, improving machine vision, artificial intelligence, and automation technology will help the development of high-performance universal grafting robots.

To solve the machining operation sequencing problem in the computer aided process planning, this paper presents a hybrid genetic algorithm and simulated annealing approach for machining operation sequencing optimization in a dynamic workshop environment. The directed graph used as an explicit constraint model is formulated based on precedence constraints among machining operations, and the graph search algorithms is embedded into framework of the optimization system. The initial solutions composed of all feasible operation sequences in GA optimization stage are produced by applying a stochastic topologic sort algorithm to the OPG. Production cost calculating model is taken as the criterion to evaluate the operation sequence quantitatively. The optimization approach can make a dynamical respond to the changes of plant resources and multiple optimal/suboptimal solutions could be obtained. Finally an illustrative example for a complicated part is given, and the test results testify the feasibility and validity of this developed method.

Grafting is a widely used technique in watermelon (Citrullus lanatus) production. However, cost of grafted seedlings is generally high as a result of intensive labor inputs for propagation using traditional grafting methods such as the manual removal of rootstock regrowth. This study developed a new grafting tool to physically remove the epidermis of pumpkin (C. maxima × C. moschata) and bottle gourd (Lagenaria siceraria) rootstock cotyledon base during grafting; we called this a new grafting method. Compared with the traditional grafting, the new grafting method significantly decreased the pumpkin rootstock regrowth rate from 100% to 8% in hole insertion and 2% in one cotyledon grafting, respectively. These attenuated rates for bottle gourd rootstock regrowth were 23% and 9% in hole insertion and one cotyledon grafting, respectively. The scion dry weights of new hole insertion and one cotyledon grafting were increased by 78% and 74% when pumpkin was used as rootstock as compared with traditional grafting without regrown rootstock removal, while the respective values were 33% and 17% in bottle gourd rootstock grafted plants. In addition, the time used for the new hole insertion grafting method to physically remove the epidermis of pumpkin rootstock cotyledon base was significantly shorter than the time required to remove the rootstock regrowth manually three times in the traditional grafting (4.2 s/plant vs. 9.3 s/plant). Similar results were also observed in the new one cotyledon grafting (4.2 s/plant vs. 8.8 s/plant). Taken together, this study presents a new method in watermelon grafting to reduce rootstock regrowth, therefore benefiting both scion growth and plant management, thus the development of this new method is clearly useful in watermelon production.

At present, the existing vegetable grafting machines are cutting and grafting operations for a single plant or row. They need to manually or automatically grab the seedlings, and their grafting efficiency is not significantly higher than that of manual grafting techniques. In this paper, grafted melon seedlings were the subject of the research. Based on the splice-grafting method, we designed a clamping and positioning device for full-tray seedlings and a mechanical device to realize full-tray grafting by locating the vegetable seedlings and completing the full-tray grafting process without damaging the seedlings. The results show that the cutting success rate for pumpkin rootstock and for melon scion reached 100% and 92%, respectively. The lengths of the long axis of the section of the rootstock and the scion were 6.2–7.7 mm, and the cutting angle was maintained at 30 degrees, thereby able to meet the requirements of the grafting method. The average grafting efficiency for the rootstock and scion were 2134 plants/h, and the average grafting success rate was 67%.

Egg size is a crucial indicator for consumer evaluation and quality grading. The main goal of this study is to measure eggs’ major and minor axes based on deep learning and single-view metrology. In this paper, we designed an egg-carrying component to obtain the actual outline of eggs. The Segformer algorithm was used to segment egg images in small batches. This study proposes a single-view measurement method suitable for eggs. Experimental results verified that the Segformer could obtain high segmentation accuracy for egg images in small batches. The mean intersection over union of the segmentation model was 96.15%, and the mean pixel accuracy was 97.17%. The R-squared was 0.969 (for the long axis) and 0.926 (for the short axis), obtained through the egg single-view measurement method proposed in this paper.

In order to address the challenges of high labour intensity, low efficiency and harsh working environment in Chinese duck egg collecting operations, this study proposes an optimization technique for egg-collecting robot path planning based on ROS (Robot Operating System). Firstly, the Gmapping algorithm is used to map the laboratory environment for navigation purposes, while the AMCL (Adaptive Monte Carlo Localization) method is used to obtain the pose information of the egg-collecting robot. Secondly，the sequence of egg collecting tasks is determined using the ACA （Ant Colony Algorithm） to compute the shortest path length and node order. Finally， the navigation target points for egg collecting are computed to obtain their world coordinate positions. The global path planning for egg collecting is performed using the Dijkstra algorithm, and the local path planning and obstacle avoidance for the robot is implemented using the DWA (Dynamic Window Approach) algorithm. A test platform was set up based on the actual working environment of the egg-collecting robot to conduct continuous egg collecting navigation experiments, and the final egg collecting navigation accuracy rate was 89.3%.