NAO humnoid robotite värvide kalibreerimise, väravate tuvastamise- ja lokaliseerimissüsteemide arendamine
Files
Date
2015
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
Selle lõputöö teemaks on autonoomsete robotite jalgpalli tarkvara arendamine.Vaatluse all on teemad nagu värvide kalibreerimine, objetkituvastus ja lokaliseerimine. Uus YUV värviruumi põhine automaatne värvide kalibreerimine on pakutud. Esitatakse detailne kirjeldus automaatse värvide kalibreerimise algoritmi implemenmtreerimisest koos visuaalsete näidetega, mis illustreerivad algoritmi toimimist. Samuti räägitakse täpsemalt muutustest, mis on implementeeritud väravate tuvastamise moodulis ja põhjustest nende muudatuste taga, andes hea ülevaate objekti tuvastamise algoritmi loogikast. Kirjeldatakse hetkel kasutatavat lokaliseerimissüsteemi ja pakutakse välja ning seletatakse lokaliseerimissüsteem parandamise tehnikat.
In this thesis, work regarding to autonomous robot soccer software development is presented. The work covers color calibration, object detection and robot localization topics. Novel YUV color space based method for the automation of color calibration is proposed. Detailed description of automatic color calibration technique implementation is provided along with the visual results illustrating performance of the method. Changes implemented to the goal detection module and motivation behind them are described in detail, providing good overview of the logic of the object recognition algorithm. Utilised localisation system is also described and, finally, the localization system enhancement technique is proposed and explained.
In this thesis, work regarding to autonomous robot soccer software development is presented. The work covers color calibration, object detection and robot localization topics. Novel YUV color space based method for the automation of color calibration is proposed. Detailed description of automatic color calibration technique implementation is provided along with the visual results illustrating performance of the method. Changes implemented to the goal detection module and motivation behind them are described in detail, providing good overview of the logic of the object recognition algorithm. Utilised localisation system is also described and, finally, the localization system enhancement technique is proposed and explained.