TR2024-097

Lagrangian Inspired Polynomial Estimator for black-box learning and control of underactuated systems

    •  Giacomuzzo, G., Cescon, R., Romeres, D., Carli, R., Dalla Libera, A., "Lagrangian Inspired Polynomial Estimator for black-box learning and control of underactuated systems", Learning for Dynamics & Control Conference (L4DC), July 2024.
      BibTeX TR2024-097 PDF
      • @inproceedings{Giacomuzzo2024jul,
      • author = {{Giacomuzzo, Giulio and Cescon, Riccardo and Romeres, Diego and Carli, Ruggero and Dalla Libera, Alberto}},
      • title = {Lagrangian Inspired Polynomial Estimator for black-box learning and control of underactuated systems},
      • booktitle = {Learning for Dynamics \& Control Conference (L4DC)},
      • year = 2024,
      • month = jul,
      • url = {https://www.merl.com/publications/TR2024-097}
      • }
  • MERL Contact:
  • Research Areas:

    Machine Learning, Robotics

Abstract:

The Lagrangian Inspired Polynomial (LIP) estimator [1] is a black-box estimator based on Gaussian Process Regression, recently presented for the inverse dynamics identification of Lagrangian systems. It relies on a novel multi-output kernel that embeds the structure of the Euler-Lagrange equation. In this work, we extend its analysis to the class of underactuated robots. First, we show that, despite being a black-box model, the LIP allows estimating kinetic and potential energies, as well as the inertial, Coriolis, and gravity components directly from the overall torque measures. Then we exploit these properties to derive a two-stage energy-based controller for the swing-up and stabilization of balancing robots. Experimental results on a simulated Pendubot confirm the feasibility of the proposed approach.