Closed-Loop Verbal Reinforcement Learning for Task-Level Robotic Planning
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , , , , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2026
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866912978942558208 |
|---|---|
| author | Plotnikov, Dmitrii Kolomiets, Iaroslav Maliukov, Dmitrii Kosenkov, Dmitrij Zinniatullina, Daniia Trandofilov, Artem Gazaryan, Georgii Bogatikov, Kirill Kozlov, Timofei Duchinskii, Igor Konenkov, Mikhail Cabrera, Miguel Altamirano Tsetserukou, Dzmitry |
| author_facet | Plotnikov, Dmitrii Kolomiets, Iaroslav Maliukov, Dmitrii Kosenkov, Dmitrij Zinniatullina, Daniia Trandofilov, Artem Gazaryan, Georgii Bogatikov, Kirill Kozlov, Timofei Duchinskii, Igor Konenkov, Mikhail Cabrera, Miguel Altamirano Tsetserukou, Dzmitry |
| contents | We propose a new Verbal Reinforcement Learning (VRL) framework for interpretable task-level planning in mobile robotic systems operating under execution uncertainty. The framework follows a closed-loop architecture that enables iterative policy improvement through interaction with the physical environment. In our framework, executable Behavior Trees are repeatedly refined by a Large Language Model actor using structured natural-language feedback produced by a Vision-Language Model critic that observes the physical robot and execution traces. Unlike conventional reinforcement learning, policy updates in VRL occur directly at the symbolic planning level, without gradient-based optimization. This enables transparent reasoning, explicit causal feedback, and human-interpretable policy evolution. We validate the proposed framework on a real mobile robot performing a multi-stage manipulation and navigation task under execution uncertainty. Experimental results show that the framework supports explainable policy improvements, closed-loop adaptation to execution failures, and reliable deployment on physical robotic systems. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2603_22169 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Closed-Loop Verbal Reinforcement Learning for Task-Level Robotic Planning Plotnikov, Dmitrii Kolomiets, Iaroslav Maliukov, Dmitrii Kosenkov, Dmitrij Zinniatullina, Daniia Trandofilov, Artem Gazaryan, Georgii Bogatikov, Kirill Kozlov, Timofei Duchinskii, Igor Konenkov, Mikhail Cabrera, Miguel Altamirano Tsetserukou, Dzmitry Robotics We propose a new Verbal Reinforcement Learning (VRL) framework for interpretable task-level planning in mobile robotic systems operating under execution uncertainty. The framework follows a closed-loop architecture that enables iterative policy improvement through interaction with the physical environment. In our framework, executable Behavior Trees are repeatedly refined by a Large Language Model actor using structured natural-language feedback produced by a Vision-Language Model critic that observes the physical robot and execution traces. Unlike conventional reinforcement learning, policy updates in VRL occur directly at the symbolic planning level, without gradient-based optimization. This enables transparent reasoning, explicit causal feedback, and human-interpretable policy evolution. We validate the proposed framework on a real mobile robot performing a multi-stage manipulation and navigation task under execution uncertainty. Experimental results show that the framework supports explainable policy improvements, closed-loop adaptation to execution failures, and reliable deployment on physical robotic systems. |
| title | Closed-Loop Verbal Reinforcement Learning for Task-Level Robotic Planning |
| topic | Robotics |
| url | https://arxiv.org/abs/2603.22169 |