Many plug seedling grippers have a complex structure and bulky volume in design. They have poor performance in holding substrate and keeping a high integrity rate. In this paper, a new multi-needle seedling gripper is proposed. Based on the elasto-plastic contact model (ECM) in the EDEM, numerous particles with different material properties are modeled to simulate real seedling substrate and to study the interaction between steel needle and growing substrate. Taking the medium integrity rate as an evaluation index, explore the influence of different needle diameters, insertion depths, and insertion and grabbing speeds on the substrate integrity through the response surface method. The optimized technical parameters of the gripper was obtained through Design-Expert. The results show that the insertion depth of needles has a significant effect on the integrity of the raising medium, and the optimal depth is 40 mm. Insertion and grasping speed and depth have an interaction effect, while other factors have no significant influence on substrate integrity. Optimization results imply that when the diameter of the needle was 3 mm, the insertion depth was 40 mm, and the insertion and grabbing speed was 1 m/s, the growing medium integrity rate was the highest, reaching 89.10%. Under these parameters, prototype tests were conducted, and the highest medium integrity rate was 87.34%, which increased by an average of 7.25% compared with existing studies. The research has reference value for improving the operation quality of the plug seedling transplanter and the design of the seedling gripper. It also shows that solving problems of needle-discrete substrate interaction based on the discrete element method is feasible.

The recognition and localization of strawberries are crucial for automated harvesting and yield prediction. This article proposes a novel RTF-YOLO (RepVgg-Triplet-FocalLoss-YOLO) network model for real-time strawberry detection. First, an efficient convolution module based on structural reparameterization is proposed. This module was integrated into the backbone and neck networks to improve the detection speed. Then, the triplet attention mechanism was embedded into the last two detection heads to enhance the network’s feature extraction for strawberries and improve the detection accuracy. Lastly, the focal loss function was utilized to enhance the model’s recognition capability for challenging strawberry targets, which thereby improves the model’s recall rate. The experimental results demonstrated that the RTF-YOLO model achieved a detection speed of 145 FPS (frames per second), a precision of 91.92%, a recall rate of 81.43%, and an $mAP$ (mean average precision) of 90.24% on the test dataset. Relative to the baseline of YOLOv5s, it showed improvements of 19%, 2.3%, 4.2%, and 3.6%, respectively. The RTF-YOLO model performed better than other mainstream models and addressed the problems of false positives and false negatives in strawberry detection caused by variations in illumination and occlusion. Furthermore, it significantly enhanced the speed of detection. The proposed model can offer technical assistance for strawberry yield estimation and automated harvesting.

The humanoid torso is crucial for the overall performance of a humanoid robot. Developing an effective humanoid spine is essential for enhancing this mechanism. This paper introduces a one-vertebral-disc unit inspired by human spine anatomy. A prototype was created using 3D-printed parts and commercially available components. Two general human-like motions are achieved using two servo motors and two pulleys, reducing the number of servo motors needed. The results indicate that a one-vertebral-disc unit can bend up to 15 degrees. The proposed mechanism functions effectively and successfully mimics human movements. It holds potential for integration into humanoid torsos, enabling efficient performance in human-like tasks in the future.