Provides the infrastructure to define and analyze the solutions of Partially Observable Markov Decision Processes (POMDP) models. The package includes pomdp-solve to solve POMDPs using a variety of exact and approximate value iteration algorithms. Smallwood and Sondik (1973)
by Hossein Kamalzadeh and Michael Hahsler
Provides an interface to pomdp-solve, a solver for Partially Observable Markov Decision Processes (POMDP) origninally written by Anthony R. Cassandra. The package enables the user to simply define all components of a POMDP model and solve the problem using the finite grid method. The package also contains functions to analyze and visualize the solutions (e.g., the optimal policy).
Initial CRAN release.