Ofer Dagan, University of Colorado Boulder; Tyler Becker, University of Colorado Boulder; Zachary Sunberg, University of Colorado Boulder
Keywords: Hypothesis, Planning, POMDPs, Sensor Tasking
Abstract:
When human operators of cyber-physical systems, such as the Space Force’s Space Surveillance Tracking (SST) network, encounter surprising behavior, they often consider multiple hypotheses that might explain it. In some cases, taking information-gathering actions such as additional measurements or control inputs given to the system can help resolve uncertainty and identify the most accurate hypothesis. The task of optimizing these actions can be formulated as a belief-space Markov decision process that we call a hypothesis-driven belief MDP. Unfortunately, this problem suffers from the curse of history, similar to a partially observable Markov decision process (POMDP). To plan in continuous domains, an agent needs to reason over countlessly many possible action-observation histories, each resulting in a different estimate of the partially observable system state. The problem is exacerbated in the hypothesis-driven context, since each action-observation pair spawns several estimates resulting from the different hypotheses. This paper considers the case in which each hypothesis corresponds to a different dynamic model in an underlying POMDP. We present a new belief MDP formulation that: (i) enables reasoning over multiple hypotheses, (ii) balances the goals of determining the (most likely) correct hypothesis and performing well in the underlying task, and (iii) can be solved with sparse tree search.
Date of Conference: September 16-19, 2025
Track: Space Domain Awareness