Development of an automatic fruit-set-reagent spraying robot for tomato plants to promote uniform fruit ripening.

摘要

The objective of this research is to develop a fruit-set-reagent spraying robot. In this paper, the accuracy of reagent spraying by the robot was examined. The robot is composed of a manipulator with three degrees of freedom, a machine vision system, a control unit, a spraying unit, a manual lift unit, and a vehicle unit. The machine vision system enables the robot to detect flower clusters by image processing, and the spraying unit enables the robot to spray the fruit-set reagent onto a targeted flower cluster. The robot is designed to operate at night. The spraying experiments described here were done in a greenhouse. The flower cluster recognition success rate obtained by image processing was influenced by the number of blooming flowers per cluster: the recognition success rate was 78% for a single blooming flower and more than 95% for clusters with three or four blooming flowers. The average of spraying success rate was 87%. The main reason for failure of spraying was obstruction of flowers by the leaves and main stem. It took an average of 12 s to spray a flower cluster, including image-processing time.