Key points are not available for this paper at this time.
In order to improve the automated vegetable-harvesting level, a robot system for truss tomato harvesting and two key technologies of picking-point recognition and end-effector design were proposed. An algorithm used the segmentation feature of the color difference 2r-g-b to recognize the truss tomato fruit and the assistant mark. According to the growth characteristics of the stem of tomato truss, the approximate fitting curve of stem and the contour of assistant mark were extracted to generate the optimal picking-point for location of tomato truss. The hardware structure of an end-effector based on flexible transmission was designed, and the function of cutting and grasping could be realized simultaneously by the end-effector. Experimental results show that the success rate for harvesting truss tomato was 88.6%, and the average execution time for picking a truss tomato was 37.2s.
Ji et al. (Tue,) studied this question.