Open Source Software and Hardware

Software framework

The software framework is a software written in Java that runs on any PC. It contains the central AI and is responsible for receiving camera data, filtering the data and communicating with the robots by calculating decision in the AI.


The electronics package includes schemas and layout of all electronics.
They are designed with Autodesk EAGLE.


CAD files and PDF drawings of all parts.
2011 files use CATIA.
2013 and 2014 files use Autodesk Inventor 2013.
2016, 2017, and 2018 files use Autodesk Inventor 2016.
2019 files use Autodesk Inventor 2019.


The code running on the microcontrollers of our robot. Projects are using Eclipse as IDE and the GNU GCC compiler.
Each archive contains a PDF file with setup instructions.

TIGERs grSim Simulator

Adapted version of the grSim simulator to work with our software. It includes automated install scripts for Linux, Visual Studio support for Windows and better ball simulation.

Autonomous Referee

Autonomous referee software for detecting rule violations (2014) and managing a full game (2016)

SSL Augmented Reality App

An Augmented Reality App that visualizes AI decisions in a live video or video file send through network in a protobuf packet.

Documentation and other files

All release files can be browsed through the buttons below.

(Seminar) Papers

This section lists papers written by Tigers team members during their studies or with a strong relation to the RoboCup domain.

Published Title Author
Aug 09, 2017 Real-Time Online Adaptive Feedforward Velocity Control for Unmanned Ground Vehicles Ommer, Nicolai Abstract Paper

Online adaptation of motion models enables autonomous robots to move more accurate in case of unknown disturbances. This paper pro-
poses a new adaptive compensation feedforward controller capable of online learning a compensation motion model without any prior nowledge to counteract non-modeled disturbance such as slippage or hardware mal-functions. The controller is able to prevent motion errors a priori and is well suited for real hardware due to high adaptation rate. It can be used in conjunction with any motion model as only motion errors are compensated. A simple interface enables quick deployment of other robot systems as demonstrated in Small Size and rescue Robot RoboCup leagues.

Jul 16, 2016 Development of an Autonomous Referee Software for the Small Size League Magel, Lukas Abstract Paper


Jun 03, 2015 Design and Implementation of Dynamic Offensive Strategies in RoboCup Small Size League Geiger, Mark Abstract Paper

The work described in this report includes the implementation of several new
modules and analysis methods within Sumatra that form a nearly completely
redesigned o offensivestrategy. One key task described in this report is the handling
of multiple robots in the offensive strategy to support and improve passing and
redirecting to build a dynamic and intelligent offensiveplaying behavior.
The newly created offensive design enables more exible and more scoring dangerous
o ensive plays. Furthermore the newly created code is more reliable and easier to

Mar 19, 2015 Position Control of an Omnidirectional Mobile Robot Ommer, Nicolai Abstract Paper

Moving robots in the RoboCup Small Size League with fast reaction and low latency is crucial for the overall performance of a team. A good movement control is important to overcome the frictions of the wheels of the omnidirectional robots. Because modeling the friction is quite complex, we applied a learning algorithm based on Relative Entropy Policy Search for learning an optimal movement policy. Experiments in the simulation and with a real robot show that we are able to learn a policy that leads us to a desired target state.

Jun 21, 2013 Verbesserung der Taktischen Spielfeldanalyse - Analyse / Auslagerung von Berechnungen auf die GPU Posovszky, Philipp Abstract Paper

Diese Studienarbeit befasst sich mit dem Thema der Analyse eines Spielfeldes des Robocup in der Small Size League. Dabei war das Ziel, die aktuelle Methoden zur Spielfeldanalyse in der Software „Sumatra“ aus Performancetechnischer und Analytischer Sicht zu verbessern. Dazu wurde die aktuelle Analyse auf die Grafikkarte ausgelagert um eine besser Performance zu erzielen. Desweitern wurde die Distanzfunktion, welche nach der Euklidischen Distanz berechnet wird, im alten Analyse-Verfahren durch die Mahlanobis-Distanz ersetzt. Dies ermöglicht es einen Miteinbezug der Geschwindigkeit eines einzelnen Roboters in das Analyseverfahren. Zusätzlich wurde noch für den Play-Entwickler einer Klasse implementiert, welche es ermöglicht Informationen aus dem Analyse-Feld abzuleiten.

Jun 03, 2013 AI Architecture and Standard Game Strategies in RoboCup SSL Ommer, Nicolai Abstract Paper

In the RoboCup SSL one of the key challenges are good strategies for coordinating all robots with the aim of scoring goals. The team TIGERS Mannheim uses a central software called Sumatra which has a good base for creating and choosing good strategies. There are, however, some drawbacks in the architecture and Sumatra is still missing some good strategy implementations. Many developers worked on the strategies and many left the team afterwards. Nobody had an overview of currently active plays and their status. With this report, the whole AI architecture and all implemented strategies, called plays and roles, were analysed and optimized. The aim was to get a complete overview of all strategies and make it easier to hold the overview within the code. Another aim was to make the development of plays better understandable by creating a documentation and clean up and document the code of existing plays. The new architecture enables more flexibility, less redundant code and better under- standability. Deprecated plays and roles were removed and all existing plays tested and documented. Several features were added to the GUI of Sumatra to help developers with testing their work.

Nov 14, 2012 Learning play finder for complex models and complex offensive play Ahrendt, U. et al Abstract Paper Presentation

This is a report written in conjunction with the lecture Knowledge Based Systems. The first part describes a learner that is able to decide on the selection of next play strategies based on the success or failure of past situations. The second part is about a certain play strategie with the aim of tricking the opponent.

Aug 07, 2012 Entwicklung und Implementierung defensiver Spielstrategien Mauelshagen, Malte Abstract Paper

This seminar paper describes basic strategies for defending plays in the RoboCup Small Size League for the Team Tigers Mannheim. The two scenarios "Keeper with one defender" and "Keeper with two defenders" will be presented.

Jun 05, 2012 Taktische Spielfeldanalyse im Robocup mittels Spielzugerkennung Steinbrecher, O., Birkenkampf, P. Abstract Paper

Beobachtungen des Teams Tigers der DHBW Mannheim bei der Roboterfußballweltmeis- terschaft im Jahr 2011 und Videoanalysen zeigten, dass selbst die besten Teams der Welt mit relativ simplen Spielzügen immer wieder zum Torerfolg kommen. Besonders der indi- rekte Schuss wird oft benutzt. Um diese Angriffe abwehren zu k ̈onnen wurde ein Verfahren geschaffen, dass es ermöglicht, indirekte Schüsse zu erkennen. Dafür wurde ein Framebuf- fer angelegt, mit dem man in die Vergangenheit schauen und analysieren kann, wie es zu einem Torerfolg bzw. -schuss gekommen ist. Mithilfe dieser Daten wurde ein exemplari- scher Spielzug geschaffen, der diese Art von Angriffen abwehren kann. Weitere defensive Spielzüge können von den Entwicklern mithilfe dieser Studienarbeit entwickelt werden.

Observations of the team Tigers from the DHBW Mannheim in the robot soccer world championship in 2011 and video analysis showed that even the best teams in the world keep scoring with relatively simple plays. Especially the indirect shot is often used. To fend off these attacks, a process was created that makes it possible to identify indirect shots. For this, a frame buffer has been created, which allows you to look into the past and analyze how a goal or shot on goal has happened. Using this data, an exemplary play were created that can fend off these types of attacks. With this study other defensive plays can be developed by the developers.

Jan 09, 2012 Taktische Spielfeldanalyse im Robocup mittels Rasterung des Spielfelds Steinbrecher, O., Birkenkampf, P. Abstract Paper

Wie beim echten Fußball geht es auch im Roboterfußball darum, mehr Tore als der Gegner zu schießen um zu gewinnen. Die Erfahrungen des Teams Tigers der DHBW Mannheim bei der Roboterfußballweltmeisterschaft im Jahr 2011 zeigten, dass dies eine große Her- ausforderung ist. Zur Unterstützung der Offensivbemühungen des Teams soll eine auto- matische Spielfeldanalyse implementiert werden. Diese soll den Entwicklern der Spielzüge neue M ̈oglichkeiten bieten, da so unter anderem freie R ̈aume erkannt werden k ̈onnen. Da- zu wird das Feld in viele Rechtecke unterteilt und jedes dieser Rechtecke bewertet. In die Bewertung des Rechtecks geht dabei der Abstand der Roboter zum jeweiligen Rechteck ein. Die so gewonnenen Daten können dann von den Entwicklern von Spielzügen genutzt werden.

As in real football, it is also in robot soccer to score more goals then your opponent to win. The experience of the team Tigers from the DHBW Mannheim in the robot soccer world championship in 2011 had shown that this is a big challenge. To support the offensive efforts of the team, an automatic field analysis should be implemented. This should give the developers of plays new options, as e.g. open spaces can be detected. For this, the field is divided into many rectangles and each of these rectangles is evaluated. In the evaluation of the rectangle, the distance of the robot to the rectangle is taken into account. The obtained data can then be used by the developers.