Proceedings

Ensuring Logic in the Fog: Sound POMDP Synthesis with LTL Objectives

Can Zhou, Yulong Gao, Pian Yu

Download BibTeX Cite

Proceedings of the Thirty-Fifth International Joint Conference on Artificial Intelligence. Main Track. Pages 4097-4105.

https://doi.org/10.24963/ijcai.2026/456

Keywords

  • Planning and Scheduling: POMDPs
  • Planning and Scheduling: Planning under uncertainty
  • Planning and Scheduling: Planning with Incomplete Information
  • Uncertainty in AI: Sequential decision making
  • Knowledge Representation and Reasoning: Reasoning about actions

Abstract

Synthesising autonomous agents that can navigate uncertain environments while adhering to complex temporal constraints remains a fundamental challenge. While Linear Temporal Logic (LTL) provides a rigorous language for specifying such tasks, the inherent undecidability of qualitatively verifying LTL satisfaction in partially observable Markov decision processes renders quantitative synthesis difficult, especially when designing reliable reward signals for approximate solvers. In this paper, we bridge this gap with a novel, sound reward-shaping mechanism that dynamically generates belief-dependent rewards grounded in certified LTL satisfaction. By integrating this mechanism into an enhanced Monte Carlo Planning framework, we empower agents to navigate the `fog' of partial observability with a search process focused on maximising verifiable success. Our experiments demonstrate that this approach not only thrives in scenarios where existing solvers fail but also maintains effectiveness and scalability across diverse benchmark domains.

Can Zhou, Yulong Gao, and Pian Yu. 2026. Ensuring Logic in the Fog: Sound POMDP Synthesis with LTL Objectives. In Proceedings of the Thirty-Fifth International Joint Conference on Artificial Intelligence, IJCAI-26. IJCAI, 4097–4105. https://doi.org/10.24963/ijcai.2026/456
Welcome Future Conferences AI Journal Division IJCAI Proceedings About IJCAI IJCAI Awards