Co-Learning Port-Hamiltonian Systems and Optimal Energy-Shaping Control
Ankur Kamboj, Biswadip Dey, Vaibhav Srivastava
TLDR
A new framework co-learns port-Hamiltonian models and energy-shaping controllers from data, ensuring stable, robust control.
Key contributions
- Co-learns pH system models and optimal energy-balancing passivity-based controllers (EB-PBC) from trajectory data.
- Uses alternating optimization with policy-aware data collection to refine models and controllers iteratively.
- Neural networks embed pH dynamics and EB-PBC structure, ensuring interpretability and provable stability.
- Includes dissipation regularization for robustness to sim-to-real gaps, validated on pendulum tasks.
Why it matters
This framework offers a novel way to learn stable and robust controllers for complex physical systems directly from data. By embedding physics principles, it ensures interpretability and enhances real-world applicability, addressing critical challenges in control design.
Original Abstract
We develop a physics-informed learning framework for energy-shaping control of port-Hamiltonian (pH) systems from trajectory data. The proposed approach {co-learns} a pH system model and an optimal energy-balancing passivity-based controller (EB-PBC) through alternating optimization with policy-aware data collection. At each iteration, the system model is refined using trajectory data collected under the current control policy, and the controller is re-optimized on the updated model. Both components are parameterized by neural networks that embed the pH {dynamics} and EB-PBC structure, ensuring interpretability in terms of energy {interactions}. The learned controller renders the closed-loop system inherently passive and provably stable, and exploits passive plant dynamics without canceling the natural potential. A dissipation regularization enforces strict energy decay during training, thereby enhancing robustness to sim-to-real gaps. The proposed framework is validated on state-regulation and swing-up tasks for planar and torsional pendulum systems.
📬 Weekly AI Paper Digest
Get the top 10 AI/ML arXiv papers from the week — summarized, scored, and delivered to your inbox every Monday.