Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2026
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2602.08963 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866918328845467648 |
|---|---|
| author | Friedl, Katharina Jaquier, Noémie Kim, Seungyeon Lundell, Jens Kragic, Danica |
| author_facet | Friedl, Katharina Jaquier, Noémie Kim, Seungyeon Lundell, Jens Kragic, Danica |
| contents | Model-based controllers can offer strong guarantees on stability and convergence by relying on physically accurate dynamic models. However, these are rarely available for high-dimensional mechanical systems such as deformable objects or soft robots. While neural architectures can learn to approximate complex dynamics, they are either limited to low-dimensional systems or provide only limited formal control guarantees due to a lack of embedded physical structure. This paper introduces a latent control framework based on learned structure-preserving reduced-order dynamics for high-dimensional Lagrangian systems. We derive a reduced tracking law for fully actuated systems and adopt a Riemannian perspective on projection-based model-order reduction to study the resulting latent and projected closed-loop dynamics. By quantifying the sources of modeling error, we derive interpretable conditions for stability and convergence. We extend the proposed controller and analysis to underactuated systems by introducing learned actuation patterns. Experimental results on simulated and real-world systems validate our theoretical investigation and the accuracy of our controllers. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2602_08963 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Reduced-order Control and Geometric Structure of Learned Lagrangian Latent Dynamics Friedl, Katharina Jaquier, Noémie Kim, Seungyeon Lundell, Jens Kragic, Danica Robotics Optimization and Control Model-based controllers can offer strong guarantees on stability and convergence by relying on physically accurate dynamic models. However, these are rarely available for high-dimensional mechanical systems such as deformable objects or soft robots. While neural architectures can learn to approximate complex dynamics, they are either limited to low-dimensional systems or provide only limited formal control guarantees due to a lack of embedded physical structure. This paper introduces a latent control framework based on learned structure-preserving reduced-order dynamics for high-dimensional Lagrangian systems. We derive a reduced tracking law for fully actuated systems and adopt a Riemannian perspective on projection-based model-order reduction to study the resulting latent and projected closed-loop dynamics. By quantifying the sources of modeling error, we derive interpretable conditions for stability and convergence. We extend the proposed controller and analysis to underactuated systems by introducing learned actuation patterns. Experimental results on simulated and real-world systems validate our theoretical investigation and the accuracy of our controllers. |
| title | Reduced-order Control and Geometric Structure of Learned Lagrangian Latent Dynamics |
| topic | Robotics Optimization and Control |
| url | https://arxiv.org/abs/2602.08963 |