Solving POMDPs online through HTN Planning and Monte Carlo Tree Search

This paper describes our ShoppingSpree HTN algorithm for online planning in Partially Observable Markov Decision Processes (POMDPs).  ShoppingSpree combines the HTN planning algorithm from SHOP3, extensions to SHOP3's representation to handle partial observability, and Monte Carlo Tree Search for efficient sampling in the problem space.  This paper presents only the algorithm and initial notes on the implementation: this is work in progress.