Wheeled Lab: Modern Sim2Real for Low-cost, Open-source Wheeled Robotics

Tyler Han, Preet Shah, Sidharth Rajagopal, Yanda Bao, Sanghun Jung, Sidharth Talia, Gabriel Guo, Bryan Xu, Bhaumik Mehta, Emma Romig, Rosario Scalise, Byron Boots
Proceedings of The 9th Conference on Robot Learning, PMLR 305:906-923, 2025.

Abstract

Simulation has been pivotal in recent robotics milestones and is poised to play a prominent role in the field’s future. However, recent robotic advances often rely on expensive and high-maintenance platforms, limiting access to broader robotics audiences. This work introduces Wheeled Lab, a framework for integrating the low-cost, open-source wheeled platforms that are already widely established in education and research with Isaac Lab, an open-source, widely adopted, and rapidly growing simulation framework for robotics research. Wheeled Lab thus introduces to new user communities modern techniques in Sim2Real, such as domain randomization, sensor simulation, and end-to-end learning. To kickstart educational uses, we demonstrate three state-of-the-art policies for small-scale RC cars: controlled drifting, elevation traversal, and visual navigation, each trained and deployed through zero-shot reinforcement learning. By bridging the gap between advanced Sim2Real methods and affordable, available robotics, Wheeled Lab aims to democratize access to cutting-edge tools, fostering innovation and education in a broader robotics context. The full stack, from hardware to software, is low cost and open-source.

Cite this Paper

BibTeX
@InProceedings{pmlr-v305-han25a, title = {Wheeled Lab: Modern Sim2Real for Low-cost, Open-source Wheeled Robotics}, author = {Han, Tyler and Shah, Preet and Rajagopal, Sidharth and Bao, Yanda and Jung, Sanghun and Talia, Sidharth and Guo, Gabriel and Xu, Bryan and Mehta, Bhaumik and Romig, Emma and Scalise, Rosario and Boots, Byron}, booktitle = {Proceedings of The 9th Conference on Robot Learning}, pages = {906--923}, year = {2025}, editor = {Lim, Joseph and Song, Shuran and Park, Hae-Won}, volume = {305}, series = {Proceedings of Machine Learning Research}, month = {27--30 Sep}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/v305/main/assets/han25a/han25a.pdf}, url = {https://proceedings.mlr.press/v305/han25a.html}, abstract = {Simulation has been pivotal in recent robotics milestones and is poised to play a prominent role in the field’s future. However, recent robotic advances often rely on expensive and high-maintenance platforms, limiting access to broader robotics audiences. This work introduces Wheeled Lab, a framework for integrating the low-cost, open-source wheeled platforms that are already widely established in education and research with Isaac Lab, an open-source, widely adopted, and rapidly growing simulation framework for robotics research. Wheeled Lab thus introduces to new user communities modern techniques in Sim2Real, such as domain randomization, sensor simulation, and end-to-end learning. To kickstart educational uses, we demonstrate three state-of-the-art policies for small-scale RC cars: controlled drifting, elevation traversal, and visual navigation, each trained and deployed through zero-shot reinforcement learning. By bridging the gap between advanced Sim2Real methods and affordable, available robotics, Wheeled Lab aims to democratize access to cutting-edge tools, fostering innovation and education in a broader robotics context. The full stack, from hardware to software, is low cost and open-source.} }
Endnote
%0 Conference Paper %T Wheeled Lab: Modern Sim2Real for Low-cost, Open-source Wheeled Robotics %A Tyler Han %A Preet Shah %A Sidharth Rajagopal %A Yanda Bao %A Sanghun Jung %A Sidharth Talia %A Gabriel Guo %A Bryan Xu %A Bhaumik Mehta %A Emma Romig %A Rosario Scalise %A Byron Boots %B Proceedings of The 9th Conference on Robot Learning %C Proceedings of Machine Learning Research %D 2025 %E Joseph Lim %E Shuran Song %E Hae-Won Park %F pmlr-v305-han25a %I PMLR %P 906--923 %U https://proceedings.mlr.press/v305/han25a.html %V 305 %X Simulation has been pivotal in recent robotics milestones and is poised to play a prominent role in the field’s future. However, recent robotic advances often rely on expensive and high-maintenance platforms, limiting access to broader robotics audiences. This work introduces Wheeled Lab, a framework for integrating the low-cost, open-source wheeled platforms that are already widely established in education and research with Isaac Lab, an open-source, widely adopted, and rapidly growing simulation framework for robotics research. Wheeled Lab thus introduces to new user communities modern techniques in Sim2Real, such as domain randomization, sensor simulation, and end-to-end learning. To kickstart educational uses, we demonstrate three state-of-the-art policies for small-scale RC cars: controlled drifting, elevation traversal, and visual navigation, each trained and deployed through zero-shot reinforcement learning. By bridging the gap between advanced Sim2Real methods and affordable, available robotics, Wheeled Lab aims to democratize access to cutting-edge tools, fostering innovation and education in a broader robotics context. The full stack, from hardware to software, is low cost and open-source.
APA
Han, T., Shah, P., Rajagopal, S., Bao, Y., Jung, S., Talia, S., Guo, G., Xu, B., Mehta, B., Romig, E., Scalise, R. & Boots, B.. (2025). Wheeled Lab: Modern Sim2Real for Low-cost, Open-source Wheeled Robotics. Proceedings of The 9th Conference on Robot Learning, in Proceedings of Machine Learning Research 305:906-923 Available from https://proceedings.mlr.press/v305/han25a.html.

Related Material