Solving Belief-Driven Pathfinding using Monte-Carlo Tree Search


Aversa Davide Vassos Stavros
2016 DiGRA/FDG '16 - Abstract Proceedings of the First International Joint Conference of DiGRA and FDG

In this work we discuss a stochastic extension to the (discrete) Belief-Driven Pathfinding (BDP) approach for finding personalized paths based on the beliefs of a character about the current state of the map. Our stochastic BDP upgrades previous work to the more realistic setting of using probabilities for the beliefs and takes advantage of approximate Monte Carlo Tree Search approaches.