CASPIAN JOURNAL
MANAGEMENT AND HIGH TECHNOLOGIES
DESIGN OF ROBOTIC MANIPULATORS WITH COMPUTER VISION SYSTEM FOR VEGETABLES HARVESTING
 |
Read | Rybakov Alexey V., Likhter Anatoliy M., Pogozheva Alena B., Mikhailova Anastasiya V., Dusaliev Amit B. DESIGN OF ROBOTIC MANIPULATORS WITH COMPUTER VISION SYSTEM FOR VEGETABLES HARVESTING // Caspian journal : management and high technologies. — 2020. — №3. — pp. 135-147. |
Rybakov Alexey V. - Astrakhan State University, rybakov_alex@mail.ru
Likhter Anatoliy M. - Astrakhan State University, likhter@bk.ru
Pogozheva Alena B. - Astrakhan State University, goddess1806@mail.ru
Mikhailova Anastasiya V. - Astrakhan State University, 051098anastasiya@gmail.com
Dusaliev Amit B. - Astrakhan State University, amit19@mail.com
Computer vision systems have recently been used in various fields of human activity. Agriculture is no exception. In particular, various methods of automated harvesting have been developed for a long time. However, these methods mostly use some markers or specially calculated precise seedling planting. The proposed approach allows for more flexible harvesting by determining the position of the fruit based on the use of a computer vision system. The article describes an experiment on the implementation of the described approach using the KUKA educational robotic complex. We analyzed computer vision data containing the coordinates of the center of the cube; then we calculated the error values in determining the yc ordinate. The experimental results and the function of the relative error dependence on the height of the grip above the object confirm the theoretically calculated values. Based on the results obtained,a program module for positioning the manipulator was developed, in which the values of the coordinates of the center of the fetus are fed to the input, taking into account the error and the coordinates of the robot's grip; the direct and inverse problem of the manipulator's kinematics is solved in order to bring the grip to the fetus. The obtained results indicate good prospects for implementing the described technology for performing harvesting operations.
Key words: robotic manipulator, computer vision, agromechatronics, image recognition, agricultural robots, automated harvesting, KUKA KR3, robot control, OpenCV