Design and development of a machine-vision-based positioning compensation system for industrial robots

Industrial robots play a critical role in modern automated manufacturing systems, where their positioning accuracy directly influences the overall system stability and production quality. However, prolonged operation often leads to a degradation in positioning precision due to mechanical wear, thermal drift, and calibration drift. To address this issue, this study presents a machine vision-based positioning compensation system designed to enable real-time error detection and correction. The primary innovation of this work lies in the system-level integration of an industrial camera, visual processing software, a programmable logic controller (PLC), an industrial robot, and a WinCC-based human-machine interface (HMI). The camera and software form a cooperative vision module that captures workpiece images, processes them through a robust image analysis pipeline, and extracts positional deviations. The calculated compensation data are transmitted to the PLC, which orchestrates corrective actions by the robot in a closed-loop manner, thereby enhancing operational reliability. Experimental results demonstrate that the proposed integrated system achieves a positioning accuracy within 0.231 mm. The proposed vision-guided compensation framework offers a practical and effective system-level solution for enhancing the precision and robustness of industrial robotic systems in real-world manufacturing environments.