Automatic milking systems (AMSs) for medium and large dairy farms in China require manual assistance to attach the teat cup, which greatly affects the milking efficiency and labor costs. In this regard, it is necessary to realize the automatic completion of cow teat attachment work. To address this issue, the authors developed a teat cup attachment robot for an AMS based on the theory of the solution of inventive problems (TRIZ). Specifically, we developed an enhanced algorithm for teat detection and designed a six-degree-of-freedom manipulator with integrated drive control. The design parameters were simulated and analyzed to validate their efficacy, while the rationality of the manipulator’s movement during teat cup attachment was verified. The maximum displacement and angle error of the cup was 1.625 mm and 1.216 mm, respectively, as verified by the teat cup attachment error test. A dynamic response test showed that the manipulator could follow the teat of the cow in real time. The attachment time for teat cups was 21 s per cow, with a success rate of 98%. The performance of the teat cup attachment robot was capable of meeting the automatic attachment teat cup needs for medium and large dairy farms during milking.

The real-time detection and instance segmentation of strawberries constitute fundamental components in the development of strawberry harvesting robots. Real-time identification of strawberries in an unstructured environment is a challenging task. Current instance segmentation algorithms for strawberries suffer from issues such as poor real-time performance and low accuracy. To this end, the present study proposes an Efficient YOLACT (E-YOLACT) algorithm for strawberry detection and segmentation based on the YOLACT framework. The key enhancements of the E-YOLACT encompass the development of a lightweight attention mechanism, pyramid squeeze shuffle attention (PSSA), for efficient feature extraction. Additionally, an attention-guided context-feature pyramid network (AC-FPN) is employed instead of FPN to optimize the architecture’s performance. Furthermore, a feature-enhanced model (FEM) is introduced to enhance the prediction head’s capabilities, while efficient fast non-maximum suppression (EF-NMS) is devised to improve non-maximum suppression. The experimental results demonstrate that the E-YOLACT achieves a Box-mAP and Mask-mAP of 77.9 and 76.6, respectively, on the custom dataset. Moreover, it exhibits an impressive category accuracy of 93.5%. Notably, the E-YOLACT also demonstrates a remarkable real-time detection capability with a speed of 34.8 FPS. The method proposed in this article presents an efficient approach for the vision system of a strawberry-picking robot.