Control algorithm design for robotic arm visual servoing in ROS 2

This bachelor thesis focuses on the development and implementation of an image-based visual servoing algorithm for a 4 DOF desktop robotic manipulator. Designed to operate over objects positioned on a 2D plane, the system uses an eye-in-hand configuration to capture real-time image data that guides the robot's movements. An object detection model identifies and locates objects within the workspace, enabling the controller to generate precise commands for the manipulator’s actions. The entire system is developed and integrated using ROS2, providing a modular and flexible framework for robotic control. By combining precise object detection with visual servoing control theory, this project demonstrates a robust and efficient approach to autonomous object manipulation, showcasing the potential of vision-based control systems in the field of intelligent robotics.