Accurate Task-Space Tracking for Humanoids with Modeling Errors Using Iterative Learning Control

Pranav A. Bhounsule, Katsu Yamane

Research output: Contribution to journalArticle

Abstract

Precise task-space tracking with manipulator-type systems requires an accurate kinematic model. In contrast to traditional manipulators, sometimes it is difficult to obtain an accurate kinematic model of humanoid robots due to complex structure and link flexibility. Also, prolonged use of the robot will lead to some parts wearing out or being replaced with a slightly different alignment, thus throwing off the initial calibration. Therefore, there is a need to develop a control algorithm that can compensate for the modeling errors and quickly retune itself, if needed, taking into account the controller bandwidth limitations and high dimensionality of the system. In this paper, we develop an iterative learning control algorithm that can work with existing inverse kinematics solvers to refine the joint-level control commands to enable precise tracking in the task space. We demonstrate the efficacy of the algorithm on a theme-park-type humanoid doing a drawing task, serving drink in a glass, and serving a drink placed on a tray without spilling. The iterative learning control algorithm is able to reduce the tracking error by at least two orders of magnitude in less than 20 trials.

Original languageEnglish (US)
Article number1750015
JournalInternational Journal of Humanoid Robotics
Volume14
Issue number3
DOIs
StatePublished - Sep 1 2017

Fingerprint

Manipulators
Kinematics
Robots
Inverse kinematics
Level control
Calibration
Bandwidth
Glass
Controllers

Keywords

  • constraint optimization
  • humanoids
  • Iterative learning control
  • task-space tracking
  • zero-reference model

ASJC Scopus subject areas

  • Mechanical Engineering
  • Artificial Intelligence

Cite this

Accurate Task-Space Tracking for Humanoids with Modeling Errors Using Iterative Learning Control. / Bhounsule, Pranav A.; Yamane, Katsu.

In: International Journal of Humanoid Robotics, Vol. 14, No. 3, 1750015, 01.09.2017.

Research output: Contribution to journalArticle

Bhounsule, Pranav A.; Yamane, Katsu / Accurate Task-Space Tracking for Humanoids with Modeling Errors Using Iterative Learning Control.

In: International Journal of Humanoid Robotics, Vol. 14, No. 3, 1750015, 01.09.2017.

Research output: Contribution to journalArticle

@article{a941207f6f2b4ebbaf3291b44a7545a9,
title = "Accurate Task-Space Tracking for Humanoids with Modeling Errors Using Iterative Learning Control",
abstract = "Precise task-space tracking with manipulator-type systems requires an accurate kinematic model. In contrast to traditional manipulators, sometimes it is difficult to obtain an accurate kinematic model of humanoid robots due to complex structure and link flexibility. Also, prolonged use of the robot will lead to some parts wearing out or being replaced with a slightly different alignment, thus throwing off the initial calibration. Therefore, there is a need to develop a control algorithm that can compensate for the modeling errors and quickly retune itself, if needed, taking into account the controller bandwidth limitations and high dimensionality of the system. In this paper, we develop an iterative learning control algorithm that can work with existing inverse kinematics solvers to refine the joint-level control commands to enable precise tracking in the task space. We demonstrate the efficacy of the algorithm on a theme-park-type humanoid doing a drawing task, serving drink in a glass, and serving a drink placed on a tray without spilling. The iterative learning control algorithm is able to reduce the tracking error by at least two orders of magnitude in less than 20 trials.",
keywords = "constraint optimization, humanoids, Iterative learning control, task-space tracking, zero-reference model",
author = "Bhounsule, {Pranav A.} and Katsu Yamane",
year = "2017",
month = "9",
doi = "10.1142/S0219843617500153",
volume = "14",
journal = "International Journal of Humanoid Robotics",
issn = "0219-8436",
publisher = "World Scientific Publishing Co. Pte Ltd",
number = "3",

}

TY - JOUR

T1 - Accurate Task-Space Tracking for Humanoids with Modeling Errors Using Iterative Learning Control

AU - Bhounsule,Pranav A.

AU - Yamane,Katsu

PY - 2017/9/1

Y1 - 2017/9/1

N2 - Precise task-space tracking with manipulator-type systems requires an accurate kinematic model. In contrast to traditional manipulators, sometimes it is difficult to obtain an accurate kinematic model of humanoid robots due to complex structure and link flexibility. Also, prolonged use of the robot will lead to some parts wearing out or being replaced with a slightly different alignment, thus throwing off the initial calibration. Therefore, there is a need to develop a control algorithm that can compensate for the modeling errors and quickly retune itself, if needed, taking into account the controller bandwidth limitations and high dimensionality of the system. In this paper, we develop an iterative learning control algorithm that can work with existing inverse kinematics solvers to refine the joint-level control commands to enable precise tracking in the task space. We demonstrate the efficacy of the algorithm on a theme-park-type humanoid doing a drawing task, serving drink in a glass, and serving a drink placed on a tray without spilling. The iterative learning control algorithm is able to reduce the tracking error by at least two orders of magnitude in less than 20 trials.

AB - Precise task-space tracking with manipulator-type systems requires an accurate kinematic model. In contrast to traditional manipulators, sometimes it is difficult to obtain an accurate kinematic model of humanoid robots due to complex structure and link flexibility. Also, prolonged use of the robot will lead to some parts wearing out or being replaced with a slightly different alignment, thus throwing off the initial calibration. Therefore, there is a need to develop a control algorithm that can compensate for the modeling errors and quickly retune itself, if needed, taking into account the controller bandwidth limitations and high dimensionality of the system. In this paper, we develop an iterative learning control algorithm that can work with existing inverse kinematics solvers to refine the joint-level control commands to enable precise tracking in the task space. We demonstrate the efficacy of the algorithm on a theme-park-type humanoid doing a drawing task, serving drink in a glass, and serving a drink placed on a tray without spilling. The iterative learning control algorithm is able to reduce the tracking error by at least two orders of magnitude in less than 20 trials.

KW - constraint optimization

KW - humanoids

KW - Iterative learning control

KW - task-space tracking

KW - zero-reference model

UR - http://www.scopus.com/inward/record.url?scp=85028063205&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85028063205&partnerID=8YFLogxK

U2 - 10.1142/S0219843617500153

DO - 10.1142/S0219843617500153

M3 - Article

VL - 14

JO - International Journal of Humanoid Robotics

T2 - International Journal of Humanoid Robotics

JF - International Journal of Humanoid Robotics

SN - 0219-8436

IS - 3

M1 - 1750015

ER -