The project contains the GPT planner for reward-based discounted POMDPs.

The GPT planner implements the RTDP-Bel algorithm which is a version of RTDP designed for Goal POMDPs. Reward-based and discounted POMDPs are converted into Goal POMDPs using the transformation described in the paper:

• B. Bonet and H. Geffner. Solving POMDPs: RTDP-Bel vs. Point-based Algorithms. Proc. of 21st Int. Joint Conf. on Artificial Intelligence (IJCAI). 2009. Pages 1641-1646. (PDF)

Notes:

• For obtaining the code, go to [http:source/checkout Source].