Object Tracking For Four-Degree Robotic Arm Through Aruco Marker-Based Computer Vision

Abstract

In modern robotic applications, particularly in flexible manufacturing and human-robotcollaboration, the demand for autonomous and adaptive systems has grown substan tially [1]. One of the fundamental challenges lies in enabling a robotic arm to accuratelytrack and manipulate dynamic targets using real-time visual feedback. This becomeseven more critical in semi-structured or dynamic environments where object positionsnare not fixed and traditional hard-coded trajectories fail [2, 3]. The specific challengenaddressed in this work is to develop a computer vision-guided control system for a 4-nDOF robotic arm that can visually detect and track an object marked with an ArUco marker, and follow it accurately using real-time position estimates [1,4]. This is accom plished by integrating a calibrated camera, visual detection of the marker, coordinate transformation, and control algorithms that translate visual data into joint velocities.