Learning-Based Robust Control: Unifying Exploration and Distributional Robustness for Reliable Robotics via Free Energy
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_ | 1866908871399833600 |
|---|---|
| author | Jesawada, Hozefa Russo, Giovanni Swikir, Abdalla Abu-Dakka, Fares |
| author_facet | Jesawada, Hozefa Russo, Giovanni Swikir, Abdalla Abu-Dakka, Fares |
| contents | A key challenge towards reliable robotic control is devising computational models that can both learn policies and guarantee robustness when deployed in the field. Inspired by the free energy principle in computational neuroscience, to address these challenges, we propose a model for policy computation that jointly learns environment dynamics and rewards, while ensuring robustness to epistemic uncertainties. Expounding a distributionally robust free energy principle, we propose a modification to the maximum diffusion learning framework. After explicitly characterizing robustness of our policies to epistemic uncertainties in both environment and reward, we validate their effectiveness on continuous-control benchmarks, via both simulations and real-world experiments involving manipulation with a Franka Research~3 arm. Across simulation and zero-shot deployment, our approach narrows the sim-to-real gap, and enables repeatable tabletop manipulation without task-specific fine-tuning. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2603_06831 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Learning-Based Robust Control: Unifying Exploration and Distributional Robustness for Reliable Robotics via Free Energy Jesawada, Hozefa Russo, Giovanni Swikir, Abdalla Abu-Dakka, Fares Robotics Optimization and Control A key challenge towards reliable robotic control is devising computational models that can both learn policies and guarantee robustness when deployed in the field. Inspired by the free energy principle in computational neuroscience, to address these challenges, we propose a model for policy computation that jointly learns environment dynamics and rewards, while ensuring robustness to epistemic uncertainties. Expounding a distributionally robust free energy principle, we propose a modification to the maximum diffusion learning framework. After explicitly characterizing robustness of our policies to epistemic uncertainties in both environment and reward, we validate their effectiveness on continuous-control benchmarks, via both simulations and real-world experiments involving manipulation with a Franka Research~3 arm. Across simulation and zero-shot deployment, our approach narrows the sim-to-real gap, and enables repeatable tabletop manipulation without task-specific fine-tuning. |
| title | Learning-Based Robust Control: Unifying Exploration and Distributional Robustness for Reliable Robotics via Free Energy |
| topic | Robotics Optimization and Control |
| url | https://arxiv.org/abs/2603.06831 |