Authors

Jingyue Liu, Pablo Borja, Cosimo Della Santina

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.

Citation

  • Journal: Advanced Intelligent Systems
  • Year: 2024
  • Volume: 6
  • Issue: 5
  • Pages:
  • Publisher: Wiley
  • DOI: 10.1002/aisy.202300385

BibTeX

@article{Liu_2024,
  title={{Physics‐Informed Neural Networks to Model and Control Robots: A Theoretical and Experimental Investigation}},
  volume={6},
  ISSN={2640-4567},
  DOI={10.1002/aisy.202300385},
  number={5},
  journal={Advanced Intelligent Systems},
  publisher={Wiley},
  author={Liu, Jingyue and Borja, Pablo and Della Santina, Cosimo},
  year={2024}
}

Download the bib file

References

  • Chen, A. I., Balter, M. L., Maguire, T. J. & Yarmush, M. L. Deep learning robotic guidance for autonomous vascular access. Nature Machine Intelligence vol. 2 104–115 (2020) – 10.1038/s42256-020-0148-7
  • Ichnowski, J., Avigal, Y., Satish, V. & Goldberg, K. Deep learning can accelerate grasp-optimized motion planning. Science Robotics vol. 5 (2020) – 10.1126/scirobotics.abd7710
  • Mukherjee, D., Gupta, K., Chang, L. H. & Najjaran, H. A Survey of Robot Learning Strategies for Human-Robot Collaboration in Industrial Settings. Robotics and Computer-Integrated Manufacturing vol. 73 102231 (2022) – 10.1016/j.rcim.2021.102231
  • Stella, F., Della Santina, C. & Hughes, J. How can LLMs transform the robotic design process? Nature Machine Intelligence vol. 5 561–564 (2023) – 10.1038/s42256-023-00669-7
  • Buşoniu, L., de Bruin, T., Tolić, D., Kober, J. & Palunko, I. Reinforcement learning for control: Performance, stability, and deep approximators. Annual Reviews in Control vol. 46 8–28 (2018) – 10.1016/j.arcontrol.2018.09.005
  • Wurman, P. R. et al. Outracing champion Gran Turismo drivers with deep reinforcement learning. Nature vol. 602 223–228 (2022) – 10.1038/s41586-021-04357-7
  • Zhao, W., Queralta, J. P. & Westerlund, T. Sim-to-Real Transfer in Deep Reinforcement Learning for Robotics: a Survey. 2020 IEEE Symposium Series on Computational Intelligence (SSCI) 737–744 (2020) doi:10.1109/ssci47803.2020.9308468 – 10.1109/ssci47803.2020.9308468
  • Kulkarni, P., Kober, J., Babuška, R. & Della Santina, C. Learning Assembly Tasks in a Few Minutes by Combining Impedance Control and Residual Recurrent Reinforcement Learning. Advanced Intelligent Systems vol. 4 (2021) – 10.1002/aisy.202100095
  • Sünderhauf, N. et al. The limits and potentials of deep learning for robotics. The International Journal of Robotics Research vol. 37 405–420 (2018) – 10.1177/0278364918770733
  • Antonelli, G., Chiaverini, S. & Di Lillo, P. On data-driven identification: Is automatically discovering equations of motion from data a Chimera? Nonlinear Dynamics vol. 111 6487–6498 (2022) – 10.1007/s11071-022-08192-x
  • Karniadakis, G. E. et al. Physics-informed machine learning. Nature Reviews Physics vol. 3 422–440 (2021) – 10.1038/s42254-021-00314-5
  • Chen, M. et al. Physics-augmented deep learning for high-speed electromagnetic simulation and optimization. (2021) doi:10.21203/rs.3.rs-807786/v1 – 10.21203/rs.3.rs-807786/v1
  • Huang, B. & Wang, J. Applications of Physics-Informed Neural Networks in Power Systems - A Review. IEEE Transactions on Power Systems vol. 38 572–588 (2023) – 10.1109/tpwrs.2022.3162473
  • Mao, Z., Jagtap, A. D. & Karniadakis, G. E. Physics-informed neural networks for high-speed flows. Computer Methods in Applied Mechanics and Engineering vol. 360 112789 (2020) – 10.1016/j.cma.2019.112789
  • Amini Niaki, S., Haghighat, E., Campbell, T., Poursartip, A. & Vaziri, R. Physics-informed neural network for modelling the thermochemical curing process of composite-tool systems during manufacture. Computer Methods in Applied Mechanics and Engineering vol. 384 113959 (2021) – 10.1016/j.cma.2021.113959
  • Greydanus S., Adv. Neural Inf. Process. Syst. (2019)
  • Zhong Y. D., Adv. Neural Inf. Process. Syst. (2021)
  • Bhattoo, R., Ranu, S. & Krishnan, N. M. A. Learning the dynamics of particle-based systems with Lagrangian graph neural networks. Machine Learning: Science and Technology vol. 4 015003 (2023) – 10.1088/2632-2153/acb03e
  • Roehrl, M. A., Runkler, T. A., Brandtstetter, V., Tokic, M. & Obermayer, S. Modeling System Dynamics with Physics-Informed Neural Networks Based on Lagrangian Mechanics. IFAC-PapersOnLine vol. 53 9195–9200 (2020) – 10.1016/j.ifacol.2020.12.2182
  • Zhong Y. D., Learning for Dynamics and Control (2021)
  • Bhattoo R., Adv. Neural Inf. Process. Syst. (2022)
  • Lutter, M. & Peters, J. Combining physics and deep learning to learn continuous-time dynamics models. The International Journal of Robotics Research vol. 42 83–107 (2023) – 10.1177/02783649231169492
  • Della Santina, C., Catalano, M. G. & Bicchi, A. Soft Robots. Encyclopedia of Robotics 1–14 (2020) doi:10.1007/978-3-642-41610-1_146-1 – 10.1007/978-3-642-41610-1_146-1
  • Gupta J. K., Learning for Dynamics and Control (2020)
  • Murray R. M., A Mathematical Introduction to Robotic Manipulation (1994)
  • Khalil H. K., Nonlinear Control (2015)
  • Della Santina, C., Duriez, C. & Rus, D. Model-Based Control of Soft Robots: A Survey of the State of the Art and Open Challenges. IEEE Control Systems vol. 43 30–65 (2023) – 10.1109/mcs.2023.3253419
  • Zheng, Y., Hu, C., Wang, X. & Wu, Z. Physics-informed recurrent neural network modeling for predictive control of nonlinear processes. Journal of Process Control vol. 128 103005 (2023) – 10.1016/j.jprocont.2023.103005
  • Sanyal, S. & Roy, K. RAMP-Net: A Robust Adaptive MPC for Quadrotors via Physics-informed Neural Network. 2023 IEEE International Conference on Robotics and Automation (ICRA) 1019–1025 (2023) doi:10.1109/icra48891.2023.10161410 – 10.1109/icra48891.2023.10161410
  • Hewing, L., Kabzan, J. & Zeilinger, M. N. Cautious Model Predictive Control Using Gaussian Process Regression. IEEE Transactions on Control Systems Technology vol. 28 2736–2743 (2020) – 10.1109/tcst.2019.2949757
  • Mitsioni, I., Tajvar, P., Kragic, D., Tumova, J. & Pek, C. Safe Data-Driven Model Predictive Control of Systems With Complex Dynamics. IEEE Transactions on Robotics vol. 39 3242–3258 (2023) – 10.1109/tro.2023.3266995
  • Sanchez-Escalonilla, S., Reyes-Baez, R. & Jayawardhana, B. Stabilization of Underactuated Systems of Degree One via Neural Interconnection and Damping Assignment – Passivity Based Control. 2022 IEEE 61st Conference on Decision and Control (CDC) 2463–2468 (2022) doi:10.1109/cdc51059.2022.9993241 – 10.1109/cdc51059.2022.9993241
  • Arnold, F. & King, R. State–space modeling for control based on physics-informed neural networks. Engineering Applications of Artificial Intelligence vol. 101 104195 (2021) – 10.1016/j.engappai.2021.104195
  • Mowlavi, S. & Nabi, S. Optimal control of PDEs using physics-informed neural networks. Journal of Computational Physics vol. 473 111731 (2023) – 10.1016/j.jcp.2022.111731
  • Nicodemus, J., Kneifl, J., Fehr, J. & Unger, B. Physics-informed Neural Networks-based Model Predictive Control for Multi-link Manipulators. IFAC-PapersOnLine vol. 55 331–336 (2022) – 10.1016/j.ifacol.2022.09.117
  • Lutter, M., Listmann, K. & Peters, J. Deep Lagrangian Networks for end-to-end learning of energy-based control for under-actuated systems. 2019 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) 7718–7725 (2019) doi:10.1109/iros40897.2019.8968268 – 10.1109/iros40897.2019.8968268
  • Della Santina C., Encyclopedia of Robotics (2021)
  • Laschi, C., Thuruthel, T. G., Lida, F., Merzouki, R. & Falotico, E. Learning-Based Control Strategies for Soft Robots: Theory, Achievements, and Future Challenges. IEEE Control Systems vol. 43 100–113 (2023) – 10.1109/mcs.2023.3253421
  • Trefethen L. N., Numerical Linear Algebra (2022)
  • Petersen K. B., The Matrix Cookbook (2008)
  • Tomei, P. A simple PD controller for robots with elastic joints. IEEE Transactions on Automatic Control vol. 36 1208–1213 (1991) – 10.1109/9.90238
  • Kelly, R. & Salgado, R. PD control with computed feedforward of robot manipulators: a design procedure. IEEE Transactions on Robotics and Automation vol. 10 566–571 (1994) – 10.1109/70.313108
  • Hannan, M. W. & Walker, I. D. Kinematics and the Implementation of an Elephant’s Trunk Manipulator and Other Continuum Style Robots. Journal of Robotic Systems vol. 20 45–63 (2003) – 10.1002/rob.10070
  • Della Santina, C., Bicchi, A. & Rus, D. On an Improved State Parametrization for Soft Robots With Piecewise Constant Curvature and Its Use in Model Based Control. IEEE Robotics and Automation Letters vol. 5 1001–1008 (2020) – 10.1109/lra.2020.2967269