The estimation of orchard blooming levels and the determination of peak blooming dates are very important because they determine the timing of orchard flower thinning and are essential for apple yield and quality. In this paper, we propose an orchard blooming level estimation method for global-level and block-level blooming level estimation of orchards. The method consists of a deep learning-based apple flower detector, a blooming level estimator, and a peak blooming day finding estimator. The YOLOv5s model is used as the apple flower detector, which is improved by adding a coordinate attention layer and a small object detection layer and by replacing the model neck with a bidirectional feature pyramid network (BiFPN) structure to improve the performance of the apple flower detector at different growth stages. The robustness of the apple flower detector under different light conditions and the generalization across years was tested using apple flower data collected in 2021–2022. The trained apple flower detector achieved a mean average precision of 77.5%. The blooming level estimator estimated the orchard blooming level based on the proportion of flowers detected at different growth stages. Statistical results show that the blooming level estimator follows the trend of orchard blooming levels. The peak blooming day finding estimator successfully positioned the peak blooming time and provided information for the flower thinning timing decision. The method described in this paper is able to provide orchardists with accurate information on apple flower growth status and is highly automated.

To address the challenges of manual apple harvesting and the limitations of existing devices—such as constrained workspace, low efficiency, and limited flexibility—a two-stage, sub-region harvesting device was developed. The design, informed by the fruit distribution characteristics in densely planted dwarf apple orchards, integrates a positioning mechanism and a fruit-picking mechanism, enabling multiple pickings within a single positioning operation to enhance workspace coverage. A forward kinematics model was established using the Denavit–Hartenberg (D–H) parameter method. An improved Monte Carlo simulation based on a hybrid Beta distribution estimated the maximum reachable distances of the fruit-picking reference point in the X, Y, and Z directions as 2146 mm, 2169 mm, and 2165 mm, respectively—adequately covering the target harvesting domain. Incorporating a translational axis structure further expanded the harvesting volume by 1.165 m³, a 42.40% improvement over the conventional 3R configuration. To support adaptive control, a random point–geometry fusion method was proposed to solve for joint variables based on harvesting postures, and an automatic fruit-picking control system was implemented. Experimental validation, including reference point tracking and harvesting tests, demonstrated maximum positioning errors of 1.5 mm and 2.2 mm, a fruit-picking success rate of 76.53%, and an average picking time of 7.24 s per fruit—marking a 4.6% improvement compared to existing devices reported in previous studies. This study provides a comprehensive technical framework and practical reference for advancing mechanized apple harvesting.

A two-stage apple-harvesting robot was analyzed to improve motion coordination and harvesting efficiency in dwarf and high-density orchards. Screw theory was used to formulate the forward and inverse kinematics and spatial Jacobians of the positioning and picking mechanisms, while the Lagrangian and Newton–Euler methods were used to establish their dynamic models. Three predefined joint motion-time allocation schemes were compared under the same cycle constraint, and the link-length-proportional scheme was selected for subsequent analysis and experiments because it produced comparatively smooth actuator-load responses. Numerical comparison with an ADAMS rigid-body model showed close agreement under matched assumptions. Payload analysis indicated that the translational joint was most affected by increased fruit load. Planar boundary-point tracking tests produced maximum errors of 1.7 and 1.9 mm in the Y and Z directions, respectively, while the maximum joint errors were 1.7° for revolute joints and 1.6 mm for the prismatic joint. Orchard experiments achieved 287 successful picks among 371 targets, corresponding to a pooled success rate of 77.36% (95% Wilson CI: 72.83%-81.32%), with a mean picking time of 6.94 s per fruit. These results demonstrate the feasibility of the proposed modeling and motion-parameter allocation framework under the tested orchard conditions.