A Data-efficient Neural ODE Framework for Optimal Control of Soft Manipulators

Mohammadreza Kasaei, Keyhan Kouhkiloui Babarahmati, Zhibin Li, Mohsen Khadem
Proceedings of The 7th Conference on Robot Learning, PMLR 229:2700-2713, 2023.

Abstract

This paper introduces a novel approach for modeling continuous forward kinematic models of soft continuum robots by employing Augmented Neural ODE (ANODE), a cutting-edge family of deep neural network models. To the best of our knowledge, this is the first application of ANODE in modeling soft continuum robots. This formulation introduces auxiliary dimensions, allowing the system’s states to evolve in the augmented space which provides a richer set of dynamics that the model can learn, increasing the flexibility and accuracy of the model. Our methodology achieves exceptional sample efficiency, training the continuous forward kinematic model using only 25 scattered data points. Additionally, we design and implement a fully parallel Model Predictive Path Integral (MPPI)-based controller running on a GPU, which efficiently manages a non-convex objective function. Through a set of experiments, we showed that the proposed framework (ANODE+MPPI) significantly outperforms state-of-the-art learning-based methods such as FNN and RNN in unseen-before scenarios and marginally outperforms them in seen-before scenarios.

Cite this Paper

BibTeX
@InProceedings{pmlr-v229-kasaei23a, title = {A Data-efficient Neural ODE Framework for Optimal Control of Soft Manipulators}, author = {Kasaei, Mohammadreza and Babarahmati, Keyhan Kouhkiloui and Li, Zhibin and Khadem, Mohsen}, booktitle = {Proceedings of The 7th Conference on Robot Learning}, pages = {2700--2713}, year = {2023}, editor = {Tan, Jie and Toussaint, Marc and Darvish, Kourosh}, volume = {229}, series = {Proceedings of Machine Learning Research}, month = {06--09 Nov}, publisher = {PMLR}, pdf = {https://proceedings.mlr.press/v229/kasaei23a/kasaei23a.pdf}, url = {https://proceedings.mlr.press/v229/kasaei23a.html}, abstract = {This paper introduces a novel approach for modeling continuous forward kinematic models of soft continuum robots by employing Augmented Neural ODE (ANODE), a cutting-edge family of deep neural network models. To the best of our knowledge, this is the first application of ANODE in modeling soft continuum robots. This formulation introduces auxiliary dimensions, allowing the system’s states to evolve in the augmented space which provides a richer set of dynamics that the model can learn, increasing the flexibility and accuracy of the model. Our methodology achieves exceptional sample efficiency, training the continuous forward kinematic model using only 25 scattered data points. Additionally, we design and implement a fully parallel Model Predictive Path Integral (MPPI)-based controller running on a GPU, which efficiently manages a non-convex objective function. Through a set of experiments, we showed that the proposed framework (ANODE+MPPI) significantly outperforms state-of-the-art learning-based methods such as FNN and RNN in unseen-before scenarios and marginally outperforms them in seen-before scenarios.} }
Endnote
%0 Conference Paper %T A Data-efficient Neural ODE Framework for Optimal Control of Soft Manipulators %A Mohammadreza Kasaei %A Keyhan Kouhkiloui Babarahmati %A Zhibin Li %A Mohsen Khadem %B Proceedings of The 7th Conference on Robot Learning %C Proceedings of Machine Learning Research %D 2023 %E Jie Tan %E Marc Toussaint %E Kourosh Darvish %F pmlr-v229-kasaei23a %I PMLR %P 2700--2713 %U https://proceedings.mlr.press/v229/kasaei23a.html %V 229 %X This paper introduces a novel approach for modeling continuous forward kinematic models of soft continuum robots by employing Augmented Neural ODE (ANODE), a cutting-edge family of deep neural network models. To the best of our knowledge, this is the first application of ANODE in modeling soft continuum robots. This formulation introduces auxiliary dimensions, allowing the system’s states to evolve in the augmented space which provides a richer set of dynamics that the model can learn, increasing the flexibility and accuracy of the model. Our methodology achieves exceptional sample efficiency, training the continuous forward kinematic model using only 25 scattered data points. Additionally, we design and implement a fully parallel Model Predictive Path Integral (MPPI)-based controller running on a GPU, which efficiently manages a non-convex objective function. Through a set of experiments, we showed that the proposed framework (ANODE+MPPI) significantly outperforms state-of-the-art learning-based methods such as FNN and RNN in unseen-before scenarios and marginally outperforms them in seen-before scenarios.
APA
Kasaei, M., Babarahmati, K.K., Li, Z. & Khadem, M.. (2023). A Data-efficient Neural ODE Framework for Optimal Control of Soft Manipulators. Proceedings of The 7th Conference on Robot Learning, in Proceedings of Machine Learning Research 229:2700-2713 Available from https://proceedings.mlr.press/v229/kasaei23a.html.

Related Material