Abstract
Currently, deep reinforcement learning primarily focuses on simulated environments in the field of robot control. Algorithms deployed on real robots have high platform requirements, leading to practical implementation difficulties. This paper presents an easily implementable algorithm transfer framework deployed to a trifinger robot. Firstly, we obtain well-performing policy models by various deep reinforcement learning algorithms trained on a simulated environment. Through multimodal information fusion, domain randomization and observation-action space pruning, the models are successfully transferred to the real robots. The presented framework is capable of controlling a real trifinger robot to move a randomly placed target to a specified position with the success rate of 90.74%, demonstrating the feasibility of our framework and the effectiveness of our methods.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
References
Allshire, A., MittaI, M., Lodaya, V., et al.: Transferring dexterous manipulation from GPU simulation to a remote real-world trifinger. In: 2022 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), pp. 11802–11809. IEEE (2022)
Andrychowicz, O.A.I.M., Baker, B., Chociej, M., et al.: Learning dexterous in-hand manipulation. Int. J. Rob. Res. 39(1), 3–20 (2020)
Lee, J., Hwangbo, J., Wellhausen, L., et al.: Learning quadrupedal locomotion over challenging terrain. Sci. Rob. 5(47), eabc5986 (2020)
Wüthrich, M., Widmaier, F., Grimminger, F., et al.: Trifinger: an open-source robot for learning dexterity. arXiv preprint arXiv:2008.03596 (2020)
Open-source URL for TriFinger robot. https://github.com/open-dynamic-robot-initiative
Makoviychuk, V., Wawrzyniak, L., Guo, Y., et al.: Isaac gym: High performance GPU-based physics simulation for robot learning. arxiv preprint arxiv:2108.10470 (2021)
Schulman, J., Wolski, F., Dhariwal, P., et al.: Proximal policy optimization algorithms. arXiv preprint arXiv:1707.06347 (2017)
Haarnoja, T., Zhou, A., Hartikainen, K., et al.: Soft actor-critic algorithms and applications. arXiv preprint arXiv:1812.05905 (2018)
Lillicrap, T.P., Hunt, J.J., Pritzel, A., et al.: Continuous control with deep reinforcement learning. arxiv preprint arxiv:1509.02971 (2015)
Fujimoto, S., Hoof, H., Meger, D.: Addressing function approximation error in actor-critic methods. In: International Conference on Machine Learning, pp. 1587–1596. PMLR (2018)
Hwangbo, J., Lee, J., Dosovitskiy, A., et al.: Learning agile and dynamic motor skills for legged robots. Sci. Rob. 4(26), eaau5872 (2019)
Gürtler, N., Blaes, S., Kolev, P., et al.: Benchmarking offline reinforcement learning on real-robot hardware. arxiv preprint arxiv:2307.15690 (2023)
Jocher, G., Chaurasia, A., Qiu, J.: Ultralytics YOLO (Version 8.0.0) [Computer software] (2023). https://github.com/ultralytics/ultralytics
Acknowledgments
This work was supported in part by NSFC under grant No.62125305, No. U23A20339, No.62088102, No.62203348.
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2025 The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.
About this paper
Cite this paper
Wan, Q., Wu, T., Ye, J., Wan, L., Lan, X. (2025). Sim-to-Real Control of Trifinger Robot by Deep Reinforcement Learning. In: Lan, X., Mei, X., Jiang, C., Zhao, F., Tian, Z. (eds) Intelligent Robotics and Applications. ICIRA 2024. Lecture Notes in Computer Science(), vol 15206. Springer, Singapore. https://doi.org/10.1007/978-981-96-0792-1_23
Download citation
DOI: https://doi.org/10.1007/978-981-96-0792-1_23
Published:
Publisher Name: Springer, Singapore
Print ISBN: 978-981-96-0791-4
Online ISBN: 978-981-96-0792-1
eBook Packages: Computer ScienceComputer Science (R0)