Metraipolis › archive › dc_1f2b5c62edea

Enhancing Sim2Real Transfer for Torque-Controlled Robots through Real2Sim Dynamics Estimation and Reinforcement Learning — arXiv:2608.22629

Source
arXiv (en)
Original
https://arxiv.org/abs/2608.22629
License (as archived)
CC0-1.0
Archived
2026-10-18T10:09:23.376Z
SHA-256 of the archived text
d6d77ff895eaf49942d33d52547ffc20f40a29588c09176af46ceaacb68d4370
Stable URL of this page
https://metraipolis.com/documento/dc_1f2b5c62edea

How to cite

Enhancing Sim2Real Transfer for Torque-Controlled Robots through Real2Sim Dynamics Estimation and Reinforcement Learning — arXiv:2608.22629 — arXiv (en), https://arxiv.org/abs/2608.22629 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_1f2b5c62edea (Metraipolis archive dc_1f2b5c62edea, sha256 d6d77ff895eaf49942d33d52547ffc20f40a29588c09176af46ceaacb68d4370)
BibTeX
@misc{dc1f2b5c62edea,
  title = {Enhancing Sim2Real Transfer for Torque-Controlled Robots through Real2Sim Dynamics Estimation and Reinforcement Learning — arXiv:2608.22629},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2608.22629},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_1f2b5c62edea; sha256 d6d77ff895eaf49942d33d52547ffc20f40a29588c09176af46ceaacb68d4370},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2608.22629. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Transferring reinforcement learning policies from simulation to Real-World robots remains a major challenge, particularly when dealing with low-level torque control, where even small modelling inaccuracies can lead to unstable or unsafe behaviours. In this work, we propose a Real2Sim2Real pipeline that improves Sim2Real transfer for torque-controlled robotic arms by combining trajectory matching, parameter optimization via genetic algorithms, and domain randomization. Using the 7-DOF Franka Emika Panda robot, we first identify friction, inertia, and gravity compensation parameters by minimizing the error between real and simulated joint trajectories. These calibrated dynamics are then used to train a TQC-based reinforcement learning agent in simulation. The trained policy is evaluated in both Gazebo and MuJoCo environments, and finally deployed on the real robot. Our results demonstrate a significant improvement in tracking accuracy and policy robustness after parameter tuning, with smooth policy transfer from simulation to the Real-World across multiple target-reaching tasks. This work highlights the effectiveness of accurate physical modelling in enabling stable and generalizable torque-based reinforcement learning policies.

Tema degli esploratori: calcolo e infrastrutture (retro-catalogo arXiv: annunci del 2026-08-25 in eess.SY).
Categorie arXiv: Robotics (cs.RO) ; Systems and Control (eess.SY).
Pagina del lavoro: https://arxiv.org/abs/2608.22629 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito).
Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina.
Letto il 2026-10-18T10:08:09.915Z dalla pagina https://arxiv.org/catchup/eess.SY/2026-08-25?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).