Research on robot autonomous inspection visual perception system based on improved YOLOv11 and SLAM

This study presents the design of an autonomous inspection robot system integrating YOLO V11 (You Only Look Once version 11) and SLAM (Simultaneous Localization and Mapping) to enhance vision-based perception and navigation in industrial environments. YOLO V11 enables real-time object detection, while SLAM ensures precise localization and mapping, allowing the robot to autonomously navigate complex environments for high-accuracy inspections. The multi-line LiDAR-based SLAM algorithm outperforms traditional visual methods by reducing errors and improving map clarity, especially in edge detection. Path planning evaluations, using both global and local algorithms, confirm the robot’s ability to avoid obstacles and accurately reach target locations. Experimental results demonstrate the system's effectiveness in real-time obstacle avoidance and precise navigation, meeting the demands of modern industrial inspection. This work highlights the potential of combining advanced machine learning and navigation technologies to create intelligent, efficient autonomous systems for industrial applications.