Research map / Detection and segmentation / Open-set and few-shot detection
POMDP Planning for Object Search in Partially Unknown Environment
Yongbo Chen, Hanna Kurniawati
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| add guessed target object with updating grid world (ablation) | POMCP · Covered1 | discounted cumulative reward: 17700.2 ± 3616.6 -> 31506.9 ± 6249.7 | Improved |
| add guessed target object with updating grid world (ablation) | POMCP · Loose2 | success rate: 65% -> 95% | Improved |
| add guessed target object with updating grid world (ablation) | POMCP · Loose1 | steps: 13.5 ± 6.8 -> 7.2 ± 0.7 | Improved |
| add guessed target object with updating grid world (ablation) | POMCP · Covered1 | success rate: 65% -> 80% | Improved |
| add guessed target object with updating grid world (ablation) | POMCP · Hidden1 | success rate: 100% -> 100% | No clear effect |
| use Monte Carlo update instead of Bellman update in GPOMCP (ablation) | Ours_b · Hidden1 | discounted cumulative reward: 45038.6 ± 7631.0 -> 55574.8 ± 6225.3 | Improved |
| use Monte Carlo update instead of Bellman update in GPOMCP (ablation) | Ours_b · Covered1 | discounted cumulative reward: 26130.6 ± 3368.7 -> 34397.8 ± 7381.6 | Improved |
| use Monte Carlo update instead of Bellman update in GPOMCP (ablation) | Ours_b · Loose2 | discounted cumulative reward: 33265.5 ± 10085.3 -> 51574.1 ± 17930.9 | Improved |
| use Monte Carlo update instead of Bellman update in GPOMCP (ablation) | Ours_b · Loose1 | discounted cumulative reward: 84701.5 ± 5546.8 -> 79788.0 ± 4787.4 | No clear effect |
| add belief tree reuse to POMCP_f (GPOMCP, Monte Carlo) (ablation) | POMCP_f · Hidden1 | discounted cumulative reward: 45815.2 ± 7260.9 -> 55574.8 ± 6225.3 | Improved |
| add belief tree reuse to POMCP_f (GPOMCP, Monte Carlo) (ablation) | POMCP_f · Covered1 | steps: 32.3 ± 6.0 -> 26.7 ± 4.8 | Improved |
| add belief tree reuse to POMCP_f (GPOMCP, Monte Carlo) (ablation) | POMCP_f · Loose1 | discounted cumulative reward: 78629.6 ± 8472.1 -> 79788.0 ± 4787.4 | No clear effect |
Closest papers by meaning
- Tru-POMDP: Task Planning Under Uncertainty via Tree of Hypotheses and Open-Ended POMDPs (NeurIPS 2025 · similarity 0.87)
- Simplifying Complex Observation Models in Continuous POMDP Planning with Probabilistic Guarantees and Practice (AAAI 2024 · similarity 0.87)
- PO-PDDL: Learning Symbolic POMDPs from Visual Demonstrations for Robot Planning Under Uncertainty (NeurIPS 2026 · similarity 0.85)
- Learning POMDP World Models from Observations with Language-Model Priors (NeurIPS 2026 · similarity 0.84)
- Multi-Environment POMDPs with Finite-Horizon Objectives (NeurIPS 2026 · similarity 0.84)
- A Surprisingly Simple Continuous-Action POMDP Solver: Lazy Cross-Entropy Search Over Policy Trees (AAAI 2024 · similarity 0.83)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
