A flat POMDP model with a known stochastic worker policy lets an online MCTS planner with goal-based reward shaping outperform plain POMCP on two simulated active goal recognition tasks.
An epsilon-optimal grid-based al- gorithm for partially observable markov deci- sion processes
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.AI 1years
2024 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Towards Intention Recognition for Robotic Assistants Through Online POMDP Planning
A flat POMDP model with a known stochastic worker policy lets an online MCTS planner with goal-based reward shaping outperform plain POMCP on two simulated active goal recognition tasks.