mGPT: A Probabilistic Planner Based on Heuristic Search

TL;DR

mGPT is a probabilistic planner that leverages heuristic search and deterministic relaxations to efficiently solve Markov Decision Processes in planning competitions.

Contribution

It introduces a novel approach combining lower bounds from deterministic relaxations with heuristic search for probabilistic planning.

Findings

Successfully participated in IPC-4 probabilistic track

Achieved efficient solving of Markov Decision Processes

Demonstrated effectiveness of heuristic search with lower bounds

Abstract

We describe the version of the GPT planner used in the probabilistic track of the 4th International Planning Competition (IPC-4). This version, called mGPT, solves Markov Decision Processes specified in the PPDDL language by extracting and using different classes of lower bounds along with various heuristic-search algorithms. The lower bounds are extracted from deterministic relaxations where the alternative probabilistic effects of an action are mapped into different, independent, deterministic actions. The heuristic-search algorithms use these lower bounds for focusing the updates and delivering a consistent value function over all states reachable from the initial state and the greedy policy.