* Description and License

RCSSJava is a java library with some (hopefully) useful code for
interacting with the Soccer Server used in RoboCup.  This library was
created by Gregory Kuhlmann and Peter Stone at the University of Texas
at Austin.  It is made freely available under the GNU Public License.

Some of the parsing in this package is done with the ANTLR parser
generator.  For convenience, we include the binary release with
this package.  For source code and more info, visit:
http://www.antlr.org/

* Intended Use

This package was designed mainly to create trainers and coaches for
the Soccer Server.  We have used it to setup up learning scenarios for
our Keepaway players 

Keepaway Site: 

http://www.cs.utexas.edu/~AustinVilla/sim/keepaway/

Selected Publications:

Keepaway Soccer: From Machine Learning Testbed to Benchmark.
Peter Stone, Gregory Kuhlmann, Matthew E. Taylor, and Yaxin Liu.
In Itsuki Noda, Adam Jacoff, Ansgar Bredenfeld, and Yasutake Takahashi, editors, RoboCup-2005: Robot Soccer World Cup IX, Springer Verlag, Berlin, 2006. To appear.

Behavior Transfer for Value-Function-Based Reinforcement Learning. 
Matthew E. Taylor and Peter Stone. 
In The Fourth International Joint Conference on Autonomous Agents and Multiagent Systems, pp. 53\u201359, ACM Press, New York, NY, July 2005.

Reinforcement Learning for RoboCup-Soccer Keepaway.
Peter Stone, Richard S. Sutton, and Gregory Kuhlmann.
Adaptive Behavior, 2005. 

These and other keepaway papers can be found here:
http://www.cs.utexas.edu/~pstone/Papers/bib2html/

Also, it is used in our simulated coach, which won the 2005 RoboCup
Coach Competition in Osaka, Japan.

Coach Site: 

http://www.cs.utexas.edu/~AustinVilla/sim/2005/).

Publications:

The UT Austin Villa 2003 Champion Simulator Coach: A Machine Learning Approach.
Gregory Kuhlmann, Peter Stone, and Justin Lallinger. 
In Daniele Nardi, Martin Riedmiller, and Claude Sammut, editors, RoboCup-2004: Robot Soccer World Cup VIII, pp. 636\u2013644, Springer Verlag, Berlin, 2005.

Guiding a Reinforcement Learner with Natural Language Advice: Initial Results in RoboCup Soccer. 
Gregory Kuhlmann, Peter Stone, Raymond Mooney, and Jude Shavlik. 
In The AAAI-2004 Workshop on Supervisory Control of Learning and Adaptive Systems, July 2004.

These and other keepaway papers can be found here:
http://www.cs.utexas.edu/~pstone/Papers/bib2html/

This package could also be extended to be useful in creating a
java-based player.

A logplayer and keepaway trainer are included in the package.

* Compilation and Installation

This package comes compiled.  To use the library, simply add
rcssjava.jar to your CLASSPATH.

The source files are included in the rcssjava directory and its
sub-directories.  If you modify the source file and would like to
generate a new rcssjava.jar file, run "make" from the top-level (if
your system supports make).  It is also possible to add this package's
top-level directory in your CLASSPATH, to use the library without 
creating a new rcssjava.jar file.

* Running included applications

To run the trainer:
java rcssjava.trainer.Trainer
OR
java -jar rcssjava.jar

To run the logplayer:
java rcssjava.logplayer.LogPlayer

Both apps accept the "-help" option to print usage information.

* Documentation

Javadoc documentation can be found in the /docs/ sub-directory.

* Need more info?

Visit http://www.cs.utexas.edu/~AustinVilla/sim/keepaway/
Or email: kuhlmann@cs.utexas.edu


