Home /Research /Radial Search Lines Method for Estimating Soccer Robot Position Using an Omnidirectional Camera
OTHER

Radial Search Lines Method for Estimating Soccer Robot Position Using an Omnidirectional Camera

Rakasiwi Bangun Hamarsudi, Iwan Kurnianto Wibowo, Mochamad Mobed Bachtiar

Year
2020
Citations
4

Abstract

Robot Soccer is one of autonomous robots that requires an ability of recognizing position to support its performance in matches. Odometry is a method that is used to calculate movements so that the robot is able to recognize its position. However, this method has a disadvantage on risk of slips in the odometry movements so that it can causes an error in calculating the position and the risk of error will be increased if it is left unchecked. In this research, a robot position estimation system was made that aims to be able to replace or support existing systems. This system utilizes omnidirectional imagery which is produced by a camera equipped with a hyperbole mirror (omnidirectional camera) as the input to determine the position of the robot This research uses radial search lines as a marker search method and the hyperbolic mirror equation to calculate the conversion of distance in pixels to meters. This system is started from searching for markers in the form of goal poles, then estimating the distance between the robot and the marker and applying a calculation using the distance to estimate the position of the robot in the field represented by x, y as coordinates and theta representing the direction of the robot The results of this system are expected to have accuracy with a measurement of error in less than 30 cm, so that the system is capable to run independently in determining the position of the robot or supporting the existing system in correcting the estimated results of its position.

Keywords

Computer visionRobotArtificial intelligenceOmnidirectional cameraOdometryPosition (finance)Computer scienceSoccer robotOmnidirectional antennaMobile robot

Related papers

Browse all OTHER papers