Autonomous Navigation Algorithm for RoboCup RRL Maneuvering 2 Field

Abstract

The goal of this project is to develop a robot platform which can autonomously accomplish the track named Maneuvering 2 of the Real Rescue Robot League of one of the most important robotics contest of the world, RoboCup. The mentioned track is made for ground robots. Robots that compete in this track should move on rays which have a contact area that has 5 cm width with their wheels and tracks. Robots may fall when they slide left or right for 5 cm. Because of competition rules, this situation causes competitors to go to the starting point and lose the score that they've earned in this ?eld until that time. Competitors may use the color difference between rays and grounds. The robot platform which is used in this study, is built on ROS framework. Movement commands are produced using the images that are taken by multiple cameras which are on the robot. OpenCV library is used to process these images. In addition to images that are taken by camera, IMU is used in the middle platform where the pass between two rays is made. The software components that generated as a result of this study can be used in a ROS installed robot platform, which had the similar sensors on it, without making any extra development.