This article presents the design, implementation and evaluation of an object classification and manipulation system in industrial environments by integrating artificial vision and a MELFA RV-2SDB robotic manipulator. The central problem lies in the need to achieve rapid and accurate classification of objects for palletizing, while ensuring the safety of operators. To address this challenge, a machine vision system based on Logitech C920 HD Pro cameras and force and torque sensors was used on the robotic manipulator. The methodology focused on the use of object and person detection algorithms, as well as direct and inverse kinematics to calculate adaptive movements of the manipulator. The experiments covered evaluation of the system's accuracy and efficiency under various lighting and environmental conditions, as well as testing people detection and geometric shape classification. The results indicated that the system allowed precise and efficient manipulation, adapting in real time to the position and characteristics of the detected objects. The conclusions highlighted the effectiveness of the system in improving productivity and safety in collaborative industrial environments, highlighting the importance of integrating cutting-edge technologies to address automation challenges in the industry.
Sorting Line Assisted by A Robotic Manipulator and Artificial Vision with Active Safety
2024-02-21
doi:10.18196/jrc.v5i2.20327
Journal of Robotics and Control (JRC); Vol 5, No 2 (2024); 388-396 ; 2715-5072 ; 2715-5056 ; 10.18196/jrc.v5i2
Aufsatz (Zeitschrift)
Elektronische Ressource
Englisch
DDC: | 629 |
Ocular Manipulator For Robotic Vision
SPIE | 1984
|Control of a Heavy-Lift Robotic Manipulator with Pneumatic Artificial Muscles
British Library Conference Proceedings | 2012
|