Learning feedback linearization control without torque measurements

Marco Capotondi; Giulio Turrisi; Claudio Gaz; Valerio Modugno; Giuseppe Oriolo; Alessandro De Luca

doi:10.5281/zenodo.4781489

Learning feedback linearization control without torque measurements

Marco Capotondi
, Giulio Turrisi
, Claudio Gaz
, Valerio Modugno
, Giuseppe Oriolo
, Alessandro De Luca

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

Abstract

Feedback Linearization (FL) allows the best control performance in executing a desired motion task when an accurate dynamic model of a fully actuated robot is available. However, due to residual parametric uncertainties and unmodeled dynamic effects, a complete cancellation of the nonlinear dynamics by feedback is hardly achieved in practice. In this paper, we summarize a novel learning framework aimed at improving online the torque correction necessary for obtaining perfect cancellation with a FL controller, using only joint position measurements. We extend then this framework to the class of underactuated robots controlled by Partial Feedback Linearization (PFL), where we simultaneously learn a feasible trajectory satisfying the boundary conditions on the desired motion while improving the associated tracking performance.

Original language	English
Title of host publication	Published in: Allotta, Benedetto, Chiara Carrozza, Maria, Menegatti, Emanuele, and Oriolo, Giuseppe (eds.) (2020) 2020 I-RIM Conference ISBN 9788894580518 Organising Body: The Institute for Robotics and Intelligent Machines Organising Body: The Institute for Robotics and Intelligent Machines
DOIs	https://doi.org/10.5281/zenodo.4781489
Publication status	Published - 10 Dec 2020

Bibliographical note

Note: Published in: Allotta, Benedetto, Chiara Carrozza, Maria, Menegatti, Emanuele, and Oriolo, Giuseppe (eds.) (2020) 2020 I-RIM Conference ISBN 9788894580518

Organising Body: The Institute for Robotics and Intelligent Machines

Organising Body: The Institute for Robotics and Intelligent Machines

Keywords

Computer science and informatics

Access to Document

10.5281/zenodo.4781489

https://doi.org/10.5281/zenodo.4781489

Learning feedback linearization control without torque measurements
Capotondi, M., Turrisi, G., Gaz, C., Modugno, V., Oriolo, G. & De Luca, A., 10 Dec 2020.
Research output: Contribution to conference › Paper › peer-review

Cite this

Capotondi, M., Turrisi, G., Gaz, C., Modugno, V., Oriolo, G., & De Luca, A. (2020). Learning feedback linearization control without torque measurements. In Published in: Allotta, Benedetto, Chiara Carrozza, Maria, Menegatti, Emanuele, and Oriolo, Giuseppe (eds.) (2020) 2020 I-RIM Conference ISBN 9788894580518 Organising Body: The Institute for Robotics and Intelligent Machines Organising Body: The Institute for Robotics and Intelligent Machines https://doi.org/10.5281/zenodo.4781489

Capotondi, Marco ; Turrisi, Giulio ; Gaz, Claudio et al. / Learning feedback linearization control without torque measurements. Published in: Allotta, Benedetto, Chiara Carrozza, Maria, Menegatti, Emanuele, and Oriolo, Giuseppe (eds.) (2020) 2020 I-RIM Conference ISBN 9788894580518 Organising Body: The Institute for Robotics and Intelligent Machines Organising Body: The Institute for Robotics and Intelligent Machines. 2020.

@inproceedings{38119d90eeab4fcf877de688988a86b7,

title = "Learning feedback linearization control without torque measurements",

abstract = "Feedback Linearization (FL) allows the best control performance in executing a desired motion task when an accurate dynamic model of a fully actuated robot is available. However, due to residual parametric uncertainties and unmodeled dynamic effects, a complete cancellation of the nonlinear dynamics by feedback is hardly achieved in practice. In this paper, we summarize a novel learning framework aimed at improving online the torque correction necessary for obtaining perfect cancellation with a FL controller, using only joint position measurements. We extend then this framework to the class of underactuated robots controlled by Partial Feedback Linearization (PFL), where we simultaneously learn a feasible trajectory satisfying the boundary conditions on the desired motion while improving the associated tracking performance.",

keywords = "Computer science and informatics",

author = "Marco Capotondi and Giulio Turrisi and Claudio Gaz and Valerio Modugno and Giuseppe Oriolo and \{De Luca\}, Alessandro",

note = "Note: Published in: Allotta, Benedetto, Chiara Carrozza, Maria, Menegatti, Emanuele, and Oriolo, Giuseppe (eds.) (2020) 2020 I-RIM Conference ISBN 9788894580518 Organising Body: The Institute for Robotics and Intelligent Machines Organising Body: The Institute for Robotics and Intelligent Machines",

year = "2020",

month = dec,

day = "10",

doi = "10.5281/zenodo.4781489",

language = "English",

booktitle = "Published in: Allotta, Benedetto, Chiara Carrozza, Maria, Menegatti, Emanuele, and Oriolo, Giuseppe (eds.) (2020) 2020 I-RIM Conference ISBN 9788894580518 Organising Body: The Institute for Robotics and Intelligent Machines Organising Body: The Institute for Robotics and Intelligent Machines",

}

Capotondi, M, Turrisi, G, Gaz, C, Modugno, V, Oriolo, G & De Luca, A 2020, Learning feedback linearization control without torque measurements. in Published in: Allotta, Benedetto, Chiara Carrozza, Maria, Menegatti, Emanuele, and Oriolo, Giuseppe (eds.) (2020) 2020 I-RIM Conference ISBN 9788894580518 Organising Body: The Institute for Robotics and Intelligent Machines Organising Body: The Institute for Robotics and Intelligent Machines. https://doi.org/10.5281/zenodo.4781489

Learning feedback linearization control without torque measurements. / Capotondi, Marco; Turrisi, Giulio; Gaz, Claudio et al.
Published in: Allotta, Benedetto, Chiara Carrozza, Maria, Menegatti, Emanuele, and Oriolo, Giuseppe (eds.) (2020) 2020 I-RIM Conference ISBN 9788894580518 Organising Body: The Institute for Robotics and Intelligent Machines Organising Body: The Institute for Robotics and Intelligent Machines. 2020.

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - Learning feedback linearization control without torque measurements

AU - Capotondi, Marco

AU - Turrisi, Giulio

AU - Gaz, Claudio

AU - Modugno, Valerio

AU - Oriolo, Giuseppe

AU - De Luca, Alessandro

N1 - Note: Published in: Allotta, Benedetto, Chiara Carrozza, Maria, Menegatti, Emanuele, and Oriolo, Giuseppe (eds.) (2020) 2020 I-RIM Conference ISBN 9788894580518 Organising Body: The Institute for Robotics and Intelligent Machines Organising Body: The Institute for Robotics and Intelligent Machines

PY - 2020/12/10

Y1 - 2020/12/10

N2 - Feedback Linearization (FL) allows the best control performance in executing a desired motion task when an accurate dynamic model of a fully actuated robot is available. However, due to residual parametric uncertainties and unmodeled dynamic effects, a complete cancellation of the nonlinear dynamics by feedback is hardly achieved in practice. In this paper, we summarize a novel learning framework aimed at improving online the torque correction necessary for obtaining perfect cancellation with a FL controller, using only joint position measurements. We extend then this framework to the class of underactuated robots controlled by Partial Feedback Linearization (PFL), where we simultaneously learn a feasible trajectory satisfying the boundary conditions on the desired motion while improving the associated tracking performance.

AB - Feedback Linearization (FL) allows the best control performance in executing a desired motion task when an accurate dynamic model of a fully actuated robot is available. However, due to residual parametric uncertainties and unmodeled dynamic effects, a complete cancellation of the nonlinear dynamics by feedback is hardly achieved in practice. In this paper, we summarize a novel learning framework aimed at improving online the torque correction necessary for obtaining perfect cancellation with a FL controller, using only joint position measurements. We extend then this framework to the class of underactuated robots controlled by Partial Feedback Linearization (PFL), where we simultaneously learn a feasible trajectory satisfying the boundary conditions on the desired motion while improving the associated tracking performance.

KW - Computer science and informatics

U2 - 10.5281/zenodo.4781489

DO - 10.5281/zenodo.4781489

M3 - Conference contribution

BT - Published in: Allotta, Benedetto, Chiara Carrozza, Maria, Menegatti, Emanuele, and Oriolo, Giuseppe (eds.) (2020) 2020 I-RIM Conference ISBN 9788894580518 Organising Body: The Institute for Robotics and Intelligent Machines Organising Body: The Institute for Robotics and Intelligent Machines

ER -

Capotondi M, Turrisi G, Gaz C, Modugno V, Oriolo G, De Luca A. Learning feedback linearization control without torque measurements. In Published in: Allotta, Benedetto, Chiara Carrozza, Maria, Menegatti, Emanuele, and Oriolo, Giuseppe (eds.) (2020) 2020 I-RIM Conference ISBN 9788894580518 Organising Body: The Institute for Robotics and Intelligent Machines Organising Body: The Institute for Robotics and Intelligent Machines. 2020 doi: 10.5281/zenodo.4781489

Learning feedback linearization control without torque measurements

Abstract

Bibliographical note

Keywords

Access to Document

Fingerprint

Research output

Learning feedback linearization control without torque measurements

Cite this