Robotic Harvesting of Rosa Damascena Using Stereoscopic Machine Vision

Abstract

4 Abstract: In this paper, we propose a system for the automated harvest of Rosa Damascena by aid of computer vision techniques. Three dimensional positions of flowers are obtained by stereo vision technique. For harvesting flowers a four-DOF manipulator is used. Also, an end-effector is designed to harvest the flowers by cutting them. To analyze the stereoscopic error, a factorial experiment in the form of a completely randomized design with 2 factors was conducted. The first factor was the distance between cameras at the three levels of 5,10 and 20cm and the second factor was the distance between the camera and the flower at the four levels of 50,75,100 and 125cm. The analysis was done using Duncan's Multiple Range Test at the 1% level. We concluded that the increase in the distance between the cameras reduces the stereoscopic error, while the increase in the distance between the cameras and the flowers increases the error. Finally, the manipulator and the vision system were evaluated altogether. The best results were obtained when the relative distance between cameras was 100 mm. In this case, 82.22 percent of the flowers were successfully harvested.