System Design of Workpiece Automatic Cleaning Device based on YOLOv5

Résumé fourni par la source

The recognition and positioning of material baskets are key links in the automatic workpiece cleaning device. Aiming at the problems of low recognition accuracy and poor precision of traditional visual methods for material basket recognition, a control system of automatic workpiece cleaning device based on YOLOv5 was designed. The YOLOv5 detection algorithm was improved by introducing the attention mechanism and optimizing the loss function, which enhanced the attention to the target area and improved the accuracy of feature extraction, thus realizing the position recognition and coordinate acquisition of workpiece material baskets. In addition, a cleaning system with Siemens S7-1200 PLC as the control core was designed. By controlling servo motors to drive the gantry and adjust the operation of the crane, the automatic grabbing and handling of material baskets were realized, and the automatic control of the cleaning process was achieved. Meanwhile, a human-computer interaction (HMI) and monitoring interface was designed, which could intuitively display the operating status of material baskets and improve the interaction capability of the automatic workpiece cleaning device.

Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.