Development of an Artificial Vision for a Parallel Manipulator Using Machine-to-Machine Technologies
Arailym Nussibaliyeva,
Gani Sergazin,
Gulzhamal Tursunbayeva,
Arman Uzbekbayev,
Nursultan Zhetenbayev,
Yerkebulan Nurgizat,
Balzhan Bakhtiyar,
Sandugash Orazaliyeva,
Saltanat Yussupova
Affiliations
Arailym Nussibaliyeva
Department of Electronics and Robotics, Almaty University of Power Engineering and Telecommunications named Gumarbek Daukeyev, Almaty 050013, Kazakhstan
Gani Sergazin
Academy of Logistics and Transport, Almaty 050012, Kazakhstan
Gulzhamal Tursunbayeva
Department of Information Security, Eurasian National University, Astana 10000, Kazakhstan
Arman Uzbekbayev
Research Institut of Applied Science and Technologies, Almaty 050013, Kazakhstan
Nursultan Zhetenbayev
Department of Electronics and Robotics, Almaty University of Power Engineering and Telecommunications named Gumarbek Daukeyev, Almaty 050013, Kazakhstan
Yerkebulan Nurgizat
Department of Electronics and Robotics, Almaty University of Power Engineering and Telecommunications named Gumarbek Daukeyev, Almaty 050013, Kazakhstan
Balzhan Bakhtiyar
Department of Heat Power Engineering, NCJSC S.Seifullin Kazakh Agro Technical Research University, Astana 10000, Kazakhstan
Sandugash Orazaliyeva
Department of Electronics and Robotics, Almaty University of Power Engineering and Telecommunications named Gumarbek Daukeyev, Almaty 050013, Kazakhstan
Saltanat Yussupova
Department of Electronics and Robotics, Almaty University of Power Engineering and Telecommunications named Gumarbek Daukeyev, Almaty 050013, Kazakhstan
This research focuses on developing an artificial vision system for a flexible delta robot manipulator and integrating it with machine-to-machine (M2M) communication to optimize real-time device interaction. This integration aims to increase the speed of the robotic system and improve its overall performance. The proposed combination of an artificial vision system with M2M communication can detect and recognize targets with high accuracy in real time within the limited space considered for positioning, further localization, and carrying out manufacturing processes such as assembly or sorting of parts. In this study, RGB images are used as input data for the MASK-R-CNN algorithm, and the results are processed according to the features of the delta robot arm prototype. The data obtained from MASK-R-CNN are adapted for use in the delta robot control system, considering its unique characteristics and positioning requirements. M2M technology enables the robot arm to react quickly to changes, such as moving objects or changes in their position, which is crucial for sorting and packing tasks. The system was tested under near real-world conditions to evaluate its performance and reliability.
