PERANCANGAN SISTEM DETEKSI OBJEK PADA ROBOT KRSBI BERBASIS MINI PC RASPBERRY PI 3

Abstract

KRSBI Wheeled is One of the competitions on the Indonesian Robot Contest,.  It is a football match that plays 3 robot  full autonomous versus other teams. The robot uses a drive in the form of wheels that are controlled in such a way, to be able to do the work the robot uses a camera sensor mounted on the front of the robot, while  for movement in the paper author uses 3 omni wheel so the robot can move in all directions to make it easier  towards the ball object. For the purposes of image processing and input and output processing the author uses a Single Board Computer Raspberry PI 3 are programmed using the Python programming language with OpenCV image processing library,  to optimize the work of Single Board Computer(SBC) Raspberry PI 3 Mini PC assisted by the Microcontroller Arduino Mega 2560. Both devices are connected serially via the USB port. Raspberry PI will process the image data obtained  webcam camera input. Next, If the ball object can be detected the object's position coordinates will be encoded in character and sent to the Microcontroller Arduino Mega 2560. Furthermore, Arduino mega 2560 will process data to drive the motors so that can move towards the position of the ball object. Based on the data from the maximum distance test results that can be read by the camera sensor to be able to detect a ball object is ±5 meters with a maximum viewing angle of 120 °.