TamedPUMA: safe and stable imitation learning with geometric fabrics

Saray Bakker, Rodrigo Perez-Dattari, Cosimo Della Santina, Wendelin Böhmer, Javier Alonso-Mora
Proceedings of the 7th Annual Learning for Dynamics \& Control Conference, PMLR 283:405-418, 2025.

Abstract

Using the language of dynamical systems, Imitation learning (IL) provides an intuitive and effective way of teaching stable task-space motions to robots with goal convergence. Yet, IL techniques are affected by serious limitations when it comes to ensuring safety and fulfillment of physical constraints. With this work, we solve this challenge via TamedPUMA, an IL algorithm augmented with a recent development in motion generation called geometric fabrics. As both the IL policy and geometric fabrics describe motions as artificial second-order dynamical systems, we propose two variations where IL provides a navigation policy for geometric fabrics. The result is a stable imitation learning strategy within which we can seamlessly blend geometrical constraints like collision avoidance and joint limits. Beyond providing a theoretical analysis, we demonstrate TamedPUMA with simulated and real-world tasks, including a 7-DoF manipulator.

Cite this Paper

BibTeX
@InProceedings{pmlr-v283-bakker25a, title = {TamedPUMA: safe and stable imitation learning with geometric fabrics}, author = {Bakker, Saray and Perez-Dattari, Rodrigo and Santina, Cosimo Della and B{\"o}hmer, Wendelin and Alonso-Mora, Javier}, booktitle = {Proceedings of the 7th Annual Learning for Dynamics \& Control Conference}, pages = {405--418}, year = {2025}, editor = {Ozay, Necmiye and Balzano, Laura and Panagou, Dimitra and Abate, Alessandro}, volume = {283}, series = {Proceedings of Machine Learning Research}, month = {04--06 Jun}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/v283/main/assets/bakker25a/bakker25a.pdf}, url = {https://proceedings.mlr.press/v283/bakker25a.html}, abstract = {Using the language of dynamical systems, Imitation learning (IL) provides an intuitive and effective way of teaching stable task-space motions to robots with goal convergence. Yet, IL techniques are affected by serious limitations when it comes to ensuring safety and fulfillment of physical constraints. With this work, we solve this challenge via TamedPUMA, an IL algorithm augmented with a recent development in motion generation called geometric fabrics. As both the IL policy and geometric fabrics describe motions as artificial second-order dynamical systems, we propose two variations where IL provides a navigation policy for geometric fabrics. The result is a stable imitation learning strategy within which we can seamlessly blend geometrical constraints like collision avoidance and joint limits. Beyond providing a theoretical analysis, we demonstrate TamedPUMA with simulated and real-world tasks, including a 7-DoF manipulator.} }
Endnote
%0 Conference Paper %T TamedPUMA: safe and stable imitation learning with geometric fabrics %A Saray Bakker %A Rodrigo Perez-Dattari %A Cosimo Della Santina %A Wendelin Böhmer %A Javier Alonso-Mora %B Proceedings of the 7th Annual Learning for Dynamics \& Control Conference %C Proceedings of Machine Learning Research %D 2025 %E Necmiye Ozay %E Laura Balzano %E Dimitra Panagou %E Alessandro Abate %F pmlr-v283-bakker25a %I PMLR %P 405--418 %U https://proceedings.mlr.press/v283/bakker25a.html %V 283 %X Using the language of dynamical systems, Imitation learning (IL) provides an intuitive and effective way of teaching stable task-space motions to robots with goal convergence. Yet, IL techniques are affected by serious limitations when it comes to ensuring safety and fulfillment of physical constraints. With this work, we solve this challenge via TamedPUMA, an IL algorithm augmented with a recent development in motion generation called geometric fabrics. As both the IL policy and geometric fabrics describe motions as artificial second-order dynamical systems, we propose two variations where IL provides a navigation policy for geometric fabrics. The result is a stable imitation learning strategy within which we can seamlessly blend geometrical constraints like collision avoidance and joint limits. Beyond providing a theoretical analysis, we demonstrate TamedPUMA with simulated and real-world tasks, including a 7-DoF manipulator.
APA
Bakker, S., Perez-Dattari, R., Santina, C.D., Böhmer, W. & Alonso-Mora, J.. (2025). TamedPUMA: safe and stable imitation learning with geometric fabrics. Proceedings of the 7th Annual Learning for Dynamics \& Control Conference, in Proceedings of Machine Learning Research 283:405-418 Available from https://proceedings.mlr.press/v283/bakker25a.html.

Related Material