Implementasi Metode Color Model Filtering HSV dan Countour Filtering untuk Mendeteksi Robot Kawan pada Robot Sepak Bola Beroda

Abstract

Wheeled soccer robot is one of the popular robotics competitions with the biggest competition being the Middle Size League. In Indonesia, the wheeled soccer robot competition is held at the Kontes Robot Indonesia (KRI). In the robot contest, the robot must be able to detect objects, one of which is a robot that is on the same team or can be called a team robot. To distinguish the team's robot from the opponent's robot, the team will be given a number background with a cyan or magenta color. The ability of the robot to detect team robot can make it easier for the robot to perform maneuvers such as movement or passing the ball. Based on this, this research tries to implement the color filtering HSV and countour filtering methods so that the robot is able to team robot. In the process of determining the threshold value, this research uses a Graphical User Interface (GUI) in order to obtain an accurate and efficient threshold value. To improve the robot's ability to detect objects, this research uses 2 camera. Camera 1 is used to detect objects in front of the robot with a large angle of 120º and camera 2 is used to detect objects with a radius of 360º. The results of this study successfully implemented the HSV color model filtering and countour filtering methods, namely on camera 1 with a detection distance of 1 meter to 8 meters and camera 2 with a radius of 1 meter to 3 meters. The data from the robot's object detection will then be sent to Arduino mega 2560 via serial communication in order to become a robot parameter in determining movement.