Monte Carlo POMDP
In the class of Markov decision process algorithms, the Monte Carlo POMDP (MC-POMDP) is the particle filter version for the partially observable Markov decision process (POMDP) algorithm. In MC-POMDP, particles filters are used to update and approximate the beliefs, and the algorithm is applicable to continuous valued states, actions, and measurements.[1]
References
- Thrun, S.; Burgard, W.; Fox, D. (2005). Probabilistic Robotics. Cambridge: The MIT Press. ISBN 0-262-20162-3.
This article is issued from Wikipedia. The text is licensed under Creative Commons - Attribution - Sharealike. Additional terms may apply for the media files.