New PO-PDDL formulation enables LLM-friendly robot planning under uncertainty

Researchers have developed PO-PDDL, a new symbolic formulation for Partially Observable Markov Decision Processes (POMDPs) that maintains the relational structure of PDDL while explicitly modeling partial observability and stochasticity. This formulation is designed to be LLM-friendly and reusable across tasks. A pipeline is proposed to learn PO-PDDL models from visual demonstrations of robot executions, identifying partial observability and learning stochastic models. Experiments on real-world manipulation tasks show that this method outperforms existing PDDL and POMDP learning approaches in robust task planning under uncertainty with reduced planning costs. AI

RANK_REASON The cluster contains an academic paper detailing a new formulation and learning pipeline for robot planning under uncertainty. [lever_c_demoted from research: ic=1 ai=1.0]

Read on arXiv cs.AI →

• arXiv
• Hugging Face
• Partially Observable Markov Decision Processes
• PO-PDDL

• paper
• model release
• other

AI-generated summary · Google Gemini · from 1 sources. How we write summaries →