Autonomous Navigation and Target Object Detection in Real-Time Mobile Robotic Welding Systems

The construction industry faces a significant shortage of skilled welders, further compounded by high safety risks associated with manual welding tasks. We propose an advanced unmanned ground vehicle (UGV) integrated with a light detection and ranging (LiDAR) sensor, visual sensors, and an autonomous robotic arm to address these challenges. Specifically, this paper presents a vision-based autonomous navigation and target object detection system for a real-time mobile robotic welding system. The proposed system can autonomously navigate complex construction sites, detect welding targets, and perform precise welding operations, thus reducing reliance on manual labor. Experimental results demonstrate the system’s capability to navigate autonomously through dynamic construction environments, achieve consistent welding quality, and effectively avoid collisions with obstacles. Future research will focus on enhancing the system’s adaptability to diverse environments, incorporating additional sensor modalities, and exploring adaptive learning algorithms to further improve autonomy and efficiency.