Abstract
This work concerns the application of physics-informed neural networks to the modeling and control of complex robotic systems. Achieving this goal requires extending physics-informed neural networks to handle nonconservative effects. These learned models are proposed to combine with model-based controllers originally developed with first-principle models in mind. By combining standard and new techniques, precise control performance can be achieved while proving theoretical stability bounds. These validations include real-world experiments of motion prediction with a soft robot and trajectory tracking with a Franka Emika Panda manipulator.
| Original language | English |
|---|---|
| Article number | 2300385 |
| Number of pages | 17 |
| Journal | Advanced Intelligent Systems |
| DOIs | |
| Publication status | Published - 2024 |
Funding
This work is supported by the EU EIC project EMERGE (grant no. 101070918). The authors are grateful to Bastian Deutschmann, the inventor of the NECK experimental platform, which greatly facilitated the work. The authors would also like to express their deepest gratitude to Francesco Stella and Tomás Coleman for their invaluable guidance and help in the experiments. Finally, the authors extend their appreciation to their colleagues for insightful feedback and constructive criticism, which helped refine the ideas and methods.Keywords
- dissipation
- Euler–Lagrange equations
- Hamiltonian neural networks
- Lagrangian neural networks
- model-based control
- physics-informed neural networks
- port-Hamiltonian systems