A Finite-State Controller Based Offline Solver for Deterministic POMDPs
Fuente:
arXiv
Gespeichert in:
| Hauptverfasser: | , , , , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2025
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866912357074075648 |
|---|---|
| author | Schutz, Alex You, Yang Mattamala, Matias Caliskanelli, Ipek Lacerda, Bruno Hawes, Nick |
| author_facet | Schutz, Alex You, Yang Mattamala, Matias Caliskanelli, Ipek Lacerda, Bruno Hawes, Nick |
| contents | Deterministic partially observable Markov decision processes (DetPOMDPs) often arise in planning problems where the agent is uncertain about its environmental state but can act and observe deterministically. In this paper, we propose DetMCVI, an adaptation of the Monte Carlo Value Iteration (MCVI) algorithm for DetPOMDPs, which builds policies in the form of finite-state controllers (FSCs). DetMCVI solves large problems with a high success rate, outperforming existing baselines for DetPOMDPs. We also verify the performance of the algorithm in a real-world mobile robot forest mapping scenario. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2505_00596 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | A Finite-State Controller Based Offline Solver for Deterministic POMDPs Schutz, Alex You, Yang Mattamala, Matias Caliskanelli, Ipek Lacerda, Bruno Hawes, Nick Robotics Artificial Intelligence Machine Learning I.2.8; I.2.9 Deterministic partially observable Markov decision processes (DetPOMDPs) often arise in planning problems where the agent is uncertain about its environmental state but can act and observe deterministically. In this paper, we propose DetMCVI, an adaptation of the Monte Carlo Value Iteration (MCVI) algorithm for DetPOMDPs, which builds policies in the form of finite-state controllers (FSCs). DetMCVI solves large problems with a high success rate, outperforming existing baselines for DetPOMDPs. We also verify the performance of the algorithm in a real-world mobile robot forest mapping scenario. |
| title | A Finite-State Controller Based Offline Solver for Deterministic POMDPs |
| topic | Robotics Artificial Intelligence Machine Learning I.2.8; I.2.9 |
| url | https://arxiv.org/abs/2505.00596 |