Multi-Hypotheses Kalman Filter based Self-Localization for autonomous soccer robots
Multi-Hypotheses Kalman Filter based Self-Localization for autonomous soccer robots
Aufgabenstellung
Autonomous soccer robots have to know where are they in the field by using perceptions, e.g. vision results and odometry. There are several methods have been proposed to solve the self-localization problem, but all of them have advantages and limits. The goal of this thesis is to implement Multi-Hypotheses Kalman Filter based Self-Localization for NAO robot in RoboCup Standard Platform League. The results will be tested in simulated and real robot. Furthermore, it will be compared with our currently employed particle filter approach.
We are going to participate in RoboCup championship 2014, the results of your thesis will be tested on the game!
DAInamite: www.dainamite.de
RoboCup SPL: www.tzi.de/spl/bin/view/Website/WebHome
Aldebaran Robotics: www.aldebaran-robotics.com/en/
Betreuer
Betreuer: Yuan Xu
Email: Yuan.Xu
dai-labor.de
Copyright TU Berlin 2007