On the Fly Adaptation of Behavior Tree-Based Policies through Reinforcement Learning
Fuente:
arXiv
Salvato in:
| Autori principali: | , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2025
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866912265760931840 |
|---|---|
| author | Iannotta, Marco Stork, Johannes A. Schaffernicht, Erik Stoyanov, Todor |
| author_facet | Iannotta, Marco Stork, Johannes A. Schaffernicht, Erik Stoyanov, Todor |
| contents | With the rising demand for flexible manufacturing, robots are increasingly expected to operate in dynamic environments where local -- such as slight offsets or size differences in workpieces -- are common. We propose to address the problem of adapting robot behaviors to these task variations with a sample-efficient hierarchical reinforcement learning approach adapting Behavior Tree (BT)-based policies. We maintain the core BT properties as an interpretable, modular framework for structuring reactive behaviors, but extend their use beyond static tasks by inherently accommodating local task variations. To show the efficiency and effectiveness of our approach, we conduct experiments both in simulation and on a Franka Emika Panda 7-DoF, with the manipulator adapting to different obstacle avoidance and pivoting tasks. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2503_06309 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | On the Fly Adaptation of Behavior Tree-Based Policies through Reinforcement Learning Iannotta, Marco Stork, Johannes A. Schaffernicht, Erik Stoyanov, Todor Robotics With the rising demand for flexible manufacturing, robots are increasingly expected to operate in dynamic environments where local -- such as slight offsets or size differences in workpieces -- are common. We propose to address the problem of adapting robot behaviors to these task variations with a sample-efficient hierarchical reinforcement learning approach adapting Behavior Tree (BT)-based policies. We maintain the core BT properties as an interpretable, modular framework for structuring reactive behaviors, but extend their use beyond static tasks by inherently accommodating local task variations. To show the efficiency and effectiveness of our approach, we conduct experiments both in simulation and on a Franka Emika Panda 7-DoF, with the manipulator adapting to different obstacle avoidance and pivoting tasks. |
| title | On the Fly Adaptation of Behavior Tree-Based Policies through Reinforcement Learning |
| topic | Robotics |
| url | https://arxiv.org/abs/2503.06309 |