Dual adaptive control of bimanual manipulation with online fuzzy parameter tuning

Alex Smith, Chenguang Yang*, Hongbin Ma, Phil Culverhouse, Angelo Cangelosi, Etienne Burdet

*Corresponding author for this work

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

5 Citations (Scopus)

Abstract

A biomimetic controller with online adaptation of impedance and force is applied to a full kinematic and dynamic model of the Baxter bimanual robot. A set of fuzzy logic engines are proposed to infer the values of tuning gains which affect the control performance and control effort of the controller, which would conventionally be set to a static value based on expert knowledge of the controller; the aim of this being to avoid the use of arbitary values to set these values. A simulated experiment is carried out, where the Baxter robot is required to move an object through a trajectory while subjected to two different disturbance forces in four phases. The controller with fuzzy inferred control gains is compared against the same controller with fixed gains to gauge the effectiveness of the new method. Results show that fuzzy inference of control gains impart an improvement in both tracking error and control effort.

Original languageEnglish
Title of host publication2014 IEEE International Symposium on Intelligent Control, ISIC 2014
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages560-565
Number of pages6
ISBN (Electronic)9781479974061
DOIs
Publication statusPublished - 25 Nov 2014
Event2014 IEEE International Symposium on Intelligent Control, ISIC 2014 - Juan Les Pins, France
Duration: 8 Oct 201410 Oct 2014

Publication series

Name2014 IEEE International Symposium on Intelligent Control, ISIC 2014

Conference

Conference2014 IEEE International Symposium on Intelligent Control, ISIC 2014
Country/TerritoryFrance
CityJuan Les Pins
Period8/10/1410/10/14

Fingerprint

Dive into the research topics of 'Dual adaptive control of bimanual manipulation with online fuzzy parameter tuning'. Together they form a unique fingerprint.

Cite this