Adaptive multivariable intermittent control: theory, development, and applications to real-time systems

Álvarez Martín, José Alberto (2018) Adaptive multivariable intermittent control: theory, development, and applications to real-time systems. PhD thesis, University of Glasgow.

Full text available as:
[thumbnail of 2018AlvarezMartinPhD.pdf] PDF
Download (26MB)
Printed Thesis Information: https://eleanor.lib.gla.ac.uk/record=b3318822

Abstract

Intermittent Control, as a control scheme that switches between open and closed-loop configurations, has been suggested as an alternative model to describe human control and to explain the intermittency observed during sustained control tasks. Additionally, IC might be beneficial in the following scenarios: 1 - in the field of robotics, where open-loop evolution could be used for computationally intensive tasks such as constrained optimisation routines, 2 - in an adaptation context, helping to detect system and environmental variations. Based on these ideas, this thesis explored the application of real-time multivariable intermittent controllers in humanoid robotics as well as adaptive versions of IC implemented on inverted pendulum structures.

Item Type: Thesis (PhD)
Qualification Level: Doctoral
Additional Information: Supported by funding from CONACyT Mexico, SCIDET Michoacan and SEP Mexico.
Keywords: Intermittent control, adaptive control, extended kalman filter, unscented kalman filter, humanoid robot, inverted pendulum, rotational pendulum, adaptation, human standing, system matched-hold, tapping hold.
Subjects: T Technology > TA Engineering (General). Civil engineering (General)
Colleges/Schools: College of Science and Engineering > School of Engineering > Biomedical Engineering
Supervisor's Name: Gollee, Dr. Henrik
Date of Award: 2018
Depositing User: Jose Alberto Alvarez Martin
Unique ID: glathesis:2018-30633
Copyright: Copyright of this thesis is held by the author.
Date Deposited: 11 Jun 2018 13:13
Last Modified: 13 Jun 2019 08:56
URI: https://theses.gla.ac.uk/id/eprint/30633

Actions (login required)

View Item View Item

Downloads

Downloads per month over past year