Based upon the amount of wins, B-Human is currently the most successful team in the RoboCup Standard platform League: They have won the World Championship six times, the German Open nine times and the European Open once. The Bremen students and their robots want to score as many goals as possible in Sydney and – as they achieved a short while ago at the German Open in Magdeburg – bring back the championship title to the Weser. In order to do this they will have to take part in probably seven games in four championship days against a total of 20 teams from 12 different countries.
All teams in the Standard Platform League play with the same hardware, namely NAO robots from the company Softbank Robotics. For the first time, a new version of the robot, the NAO V6, will be used this year. The main difference to the previous model, which can still be used, is the computing power. It is several times higher, which allows the NAOs to apply algorithms that are far more complex. The applied software varies from team to team.
New Challenges and Innovative Deep-Learning Approaches
In order to keep bridging the gap to the rules of human soccer and to promote scientific advancement, there are new challenges for the partaking teams each year. As was also the case at the German Open, this year there will also be corner kicks in Sydney, alongside free kicks and goalie kicks. Additionally, the referee will not simply place the ball back in the field and allow play to resume after a kick that causes the ball to travel over the sidelines. Rather, the opposing team will be allowed a “throw-in”, which will in fact not be a throw but a kick, a so-called “kick-in”.
The games, which are becoming ever more challenging, constantly require new algorithms adapted to the complexity. Whilst soccer was played in windowless exhibition centers under constant artificial light several years ago, it is now favored to play close to large window fronts and thus under changing light conditions. Recently, B-Human has been counting on deep learning methods in order to master the conditions that are extremely difficult for machine-based image processing. The scientific publications concerning this will be presented by the Bremen team at the RoboCup symposium, which takes place directly after the championship.
Technical Challenge and Mixed-team-Competition
Alongside the original championship games, B-Human will also take part in the so-called Technical Challenge and Mixed-Team-Competition. This year, the Technical Challenge is about correctly spatially locating the referee’s whistle as quickly as possible. The teams can also choose a technical challenge prior to the competition, which they then present on site and which is then judged by the other teams. The Bremen team will show how two NAOs lead another blindfolded team robot that has its cameras covered. In the Mixed-Team-Competition, two teams join forces in order to measure their skills against other mixed teams. B-Human is mixing with the team Berlin United from the Humboldt-Universität zu Berlin this year.
About B-Human
B-Human is made up of around 15 students from the University of Bremen as well as the supervising scientists, Dr. Thomas Röfer from the DFKI research areas of Cyber-Physical Systems, which is led by Prof. Dr. Rolf Drechsler, and Dr. Tim Laue from the University of Bremen. The company CONTACT Software, the leading supplier of solutions for product processes and the digital transformation, has been the main sponsor of B-Human since 2017.
Further information:
Official RoboCup World Cup 2019 website:
https://2019.robocup.org/index.php
B-Human team website:
https://www.b-human.de
Contact:
Dr. Tim Laue
Faculty of Mathematics and Computer Science
University of Bremen
Phone: +49 421 218-64209
Email: tlaueprotect me ?!uni-bremenprotect me ?!.de
Dr. Thomas Röfer
Deutsches Forschungszentrum für Künstliche Intelligenz GmbH (DFKI)
(German Research Center for Artificial Intelligence)
Research Area Cyber-Physical Systems
Phone: +49 421 218-64200
Email: Thomas.Roeferprotect me ?!dfkiprotect me ?!.de
