Grid-Soccer Simulator, is a grid-world environment in which two teams of agents play soccer together. The main intention behind developing this environment, is to provide a test-bed for reinforcement learning, and learning in cooperative, competitive, and mixed multi-agent environments. The communication between the environment and the clients is done through a text-based network protocol; thus, one can create a client in any desired programming language as long as that language supports networking. Currently there are several agents developed in C# and sample agents in C++ and Java, which can be easily extended to fit one's experiments' needs.
- Changes to previous version:
Initial Announcement on mloss.org.
No one has posted any comments yet. Perhaps you'd like to be the first?
Leave a comment
You must be logged in to post comments.