Data-Driven Design of Energy-Shaping Controllers for Swing-Up Control of Underactuated Robots

Wankun Sirichotiyakul, Aykut C. Satici

Research output: Chapter in Book/Report/Conference proceedingChapterpeer-review

1 Scopus citations

Abstract

We propose a novel data-driven procedure to train a neural network for the swing-up control of underactuated robotic systems. Our approach is inspired by several recent developments ranging from nonlinear control theory to machine learning. We embed a neural network indirectly into the equations of motion of the robotic manipulator as its control input. Using familiar results from passivity-based and energy-shaping control literature, this control function is determined by the appropriate gradients of a neural network, acting as an energy-like (Lyapunov) function. We encode the task of swinging-up robotic systems through the use of transverse coordinates and goal sets; which drastically accelerates the rate of learning by providing a concise target for the neural network. We demonstrate the efficacy of the algorithm with both numerical simulations and experiments.

Original languageAmerican English
Title of host publicationISER 2020: Experimental Robotics
PublisherSpringer Science and Business Media B.V.
Pages323-333
Number of pages11
DOIs
StatePublished - 2021

Publication series

NameSpringer Proceedings in Advanced Robotics
Volume19
ISSN (Print)2511-1256
ISSN (Electronic)2511-1264

EGS Disciplines

  • Biomedical Engineering and Bioengineering
  • Mechanical Engineering

Fingerprint

Dive into the research topics of 'Data-Driven Design of Energy-Shaping Controllers for Swing-Up Control of Underactuated Robots'. Together they form a unique fingerprint.

Cite this