Key points are not available for this paper at this time.
Camellia oleifera is an economically important tree species with high utilization value. At present, Camellia oleifera fruit is faced with problems such as high labor intensity and shortage of human resources through manual picking. How to realize intelligent and mechanized picking of Camellia oleifera fruit is attracting more and more attention. In this paper, a dual-track intelligent picking robot system for Camellia oleifera fruit is proposed. The system is composed of walking device, multi-axis motion platform, cooperative robot, and vision units. Firstly, the overall structure and working principle of the picking system are introduced, and the structure and design methods of the main parts such as walking device, cooperative robot, end-effector, multi-axis motion platform and dual-track are described. Secondly, a positioning method of walking device based on improved YOLOv8 is employed for the trunk of Camellia oleifera tree. A fruit picking method based on GroundingDINO visual large model is proposed for Camellia oleifera fruit recognition. Thirdly, the control unit is designed, and the control flow of the dual-track intelligent picking robot system is analyzed. Finally, the prototype of the picking robot system is completed and the prototype operation test is carried out. The results show that the designed dual-track picking system can carry a robot to complete the picking of Camellia oleifera fruit in outdoor environment, which provides a new idea for intelligent and mechanized picking of Camellia oleifera fruit.
Chen et al. (Tue,) studied this question.