Planning in Stochastic Environments with Goal Uncertainty

TL;DR

This paper introduces GUSSP, a new framework for planning in stochastic environments with goal uncertainty, extending SSP models to incorporate belief over goal states and demonstrating its effectiveness through empirical experiments.

Contribution

The paper formalizes GUSSP, a novel extension of SSP for environments with uncertain goals, and proposes heuristic and determinization methods for efficient planning.

Findings

GUSSP effectively models environments with goal uncertainty.

The proposed heuristic reduces planning time significantly.

Empirical results show successful application in search and rescue scenarios.

Abstract

We present the Goal Uncertain Stochastic Shortest Path (GUSSP) problem -- a general framework to model path planning and decision making in stochastic environments with goal uncertainty. The framework extends the stochastic shortest path (SSP) model to dynamic environments in which it is impossible to determine the exact goal states ahead of plan execution. GUSSPs introduce flexibility in goal specification by allowing a belief over possible goal configurations. The unique observations at potential goals helps the agent identify the true goal during plan execution. The partial observability is restricted to goals, facilitating the reduction to an SSP with a modified state space. We formally define a GUSSP and discuss its theoretical properties. We then propose an admissible heuristic that reduces the planning time using FLARES -- a start-of-the-art probabilistic planner. We also propose a determinization approach for solving this class of problems. Finally, we present empirical results on a search and rescue mobile robot and three other problem domains in simulation.