Can Context Bridge the Reality Gap? Sim-to-Real Transfer of Context-Aware Policies
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866908633101500416 |
|---|---|
| author | Iannotta, Marco Yang, Yuxuan Stork, Johannes A. Schaffernicht, Erik Stoyanov, Todor |
| author_facet | Iannotta, Marco Yang, Yuxuan Stork, Johannes A. Schaffernicht, Erik Stoyanov, Todor |
| contents | Sim-to-real transfer remains a major challenge in reinforcement learning (RL) for robotics, as policies trained in simulation often fail to generalize to the real world due to discrepancies in environment dynamics. Domain Randomization (DR) mitigates this issue by exposing the policy to a wide range of randomized dynamics during training, yet leading to a reduction in performance. While standard approaches typically train policies agnostic to these variations, we investigate whether sim-to-real transfer can be improved by conditioning the policy on an estimate of the dynamics parameters -- referred to as context. To this end, we integrate a context estimation module into a DR-based RL framework and systematically compare SOTA supervision strategies. We evaluate the resulting context-aware policies in both a canonical control benchmark and a real-world pushing task using a Franka Emika Panda robot. Results show that context-aware policies outperform the context-agnostic baseline across all settings, although the best supervision strategy depends on the task. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2511_04249 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Can Context Bridge the Reality Gap? Sim-to-Real Transfer of Context-Aware Policies Iannotta, Marco Yang, Yuxuan Stork, Johannes A. Schaffernicht, Erik Stoyanov, Todor Robotics Sim-to-real transfer remains a major challenge in reinforcement learning (RL) for robotics, as policies trained in simulation often fail to generalize to the real world due to discrepancies in environment dynamics. Domain Randomization (DR) mitigates this issue by exposing the policy to a wide range of randomized dynamics during training, yet leading to a reduction in performance. While standard approaches typically train policies agnostic to these variations, we investigate whether sim-to-real transfer can be improved by conditioning the policy on an estimate of the dynamics parameters -- referred to as context. To this end, we integrate a context estimation module into a DR-based RL framework and systematically compare SOTA supervision strategies. We evaluate the resulting context-aware policies in both a canonical control benchmark and a real-world pushing task using a Franka Emika Panda robot. Results show that context-aware policies outperform the context-agnostic baseline across all settings, although the best supervision strategy depends on the task. |
| title | Can Context Bridge the Reality Gap? Sim-to-Real Transfer of Context-Aware Policies |
| topic | Robotics |
| url | https://arxiv.org/abs/2511.04249 |