Vision-Based Safety System for SCARA Robots: A Real-Time Approach to Enhance Obstacle Detection

SCARA robots are widely used in industrial settings due to their speed and precision. However, ensuring workplace safety during interactions with humans and other robots is essential. This study presents a computer vision-based safety system for SCARA robots that detects obstacles in real-time. Utilizing image processing algorithms and pattern recognition techniques, the system identifies objects, such as human hands, within the robot's operational area and promptly halts its functions. The object recognition algorithm leverages the MediaPipe and OpenCV libraries in Python to detect and analyze hand movements in real-time. The safety mechanism was evaluated using a developed prototype. Experimental results demonstrate that the system effectively detects objects in complex industrial environments, thereby preventing potential accidents. This research represents an advancement in enhancing the safety of industrial robots and mitigating risks associated with human-robot interactions.

کلیدواژه ها:

SCARA Robot ، Safety Protocols ، Computer Vision ، Object Detection ، Image Processing