EnMAS: Environment for Multi-Agent Simulation is:
- a language for describing DEC-POMDPs
- a distributed DEC-POMDP engine
- a platform for artificial intelligence research
- an educational tool
The name is pronounced like en masse, which fittingly means "together" or "as one".
This project is guided by current research in the field of multi-agent machine learning, particularly the DEC-POMDP model. The current design of this project technically embraces a superset of DEC-POMDPs called Partially Observable Stochastic Games (POSGs).
- To enable rapid prototyping of POMDP problems for research and teaching purposes
- To maintain orthogonality between theoretical models and implementation
- To abstain from making assumptions about any particular problem domain
This project was created and is maintained by Connor Doyle as part of the Master of Software Engineering degree at the University of Wisconsin - La Crosse under the advisement of Drs. Marty Allen and Kenny Hunt.
The author would like to acknowledge the generosity and support of the Department of Computer Science at the University of Wisconsin - La Crosse and the National Science Foundation.
EnMAS (Environment for Multi-Agent Simulation) by Connor Doyle is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 3.0 Unported License.