Research on Machine Vision-Based Control System for Cold Storage Warehouse Robots

In recent years, the global cold chain logistics market has grown rapidly, but level of automation remains low. Compared to traditional logistics, in storage requires a balance between safety and efficiency, current detection algorithms are poor at meeting these requirements. Therefore, based on YOLOv5, this paper proposes recognition grasping system for cartons warehouses. A human–machine interaction is designed environment, enabling remote control unmanned grasping. At algorithm level, CA attention mechanism introduced improve accuracy. The Ghost lightweight module replaces CBS structure enhance runtime speed. Alpha-DIoU loss function utilized With comprehensive improvements, modified study achieves 0.711% increase mAP 0.7% FPS while maintaining Experimental results demonstrate that increases fidelity by 2.32%, reduces response time 13.89%, enhances positioning accuracy 7.14%. By incorporating all exhibits 2.16% reduction time, 4.67% improvement accuracy, significant overall performance enhancement.