Main Article Content

Abstract

Robots are a modern technology that is experiencing very rapid growth. Robots also have several types, one type of robot is Robot Vision. Robot Vision uses image processing and computer vision in carrying out its tasks, including recognizing objects and controlling robot movements. Taking the concept from the ABU (Asia Pacific Broadcasting Union) robot contest in 2024 where the robot is required to be able to pick up balls with red and blue colors which are then inserted into a place called SILO. The detection method used in this study uses the YOLO (You Only Look Once) method with the YOLOv5 version. The reason for choosing the YOLOv5 version is because of its ability to be more accurate than previous versions and lighter to run than newer versions. From the results of this study, it was found that the confidence value for each object was quite high, which was 0.90. In addition, the precision value against recall for all classes gets a value of 0.991 or 99.1%. For the ball and SILO classes, the precision to recall value is 0.993 or 99.3%. So it can be concluded that this method has good value in terms of detection and accuracy

Keywords

YOLOv5 Robocon Deteksi Objek

Article Details

References

Read More