Finite-State Controllers Based on Mealy Machines for Centralized and Decentralized POMDPs

Christopher Amato, Blai Bonet, and Shlomo Zilberstein. Finite-State Controllers Based on Mealy Machines for Centralized and Decentralized POMDPs. Proceedings of the Twenty-Fourth Conference on Artificial Intelligence (AAAI), 1052-1058, Atlanta, Georgia, 2010.

Abstract

Existing controller-based approaches for centralized and decentralized POMDPs are based on automata with output known as Moore machines. In this paper, we show that several advantages can be gained by utilizing another type of automata, the Mealy machine. Mealy machines are more powerful than Moore machines, provide a richer structure that can be exploited by solution methods, and can be easily incorporated into current controller-based approaches. To demonstrate this, we adapted some existing controller-based algorithms to use Mealy machines and obtained results on a set of benchmark domains. The Mealy-based approach always outperformed the Moore-based approach and often outperformed the state-of-the-art algorithms for both centralized and decentralized POMDPs. These findings provide fresh and general insights for the improvement of existing algorithms and the development of new ones.

Bibtex entry:

@inproceedings{ABZaaai10,
  author	= {Christopher Amato and Blai Bonet and Shlomo Zilberstein},
  title		= {Finite-State Controllers Based on Mealy Machines for Centralized
                   and Decentralized {POMDP}s},
  booktitle     = {Proceedings of the Twenty-Fourth Conference on Artificial
                   Intelligence},
  year		= {2010},
  pages		= {1052-1058},
  address       = {Atlanta, Georgia},
  url		= {http://rbr.cs.umass.edu/shlomo/papers/ABZaaai10.html}
}

shlomo@cs.umass.edu
UMass Amherst