Pushing task by repulsive compliance control in electric wheelchair using acceleration information

Tsuyoshi Shibata, Cihan Acar, Toshiyuki Murakami

Research output: Chapter in Book/Report/Conference proceedingConference contribution


This paper describes a power assist control in pushing task by an electric wheelchair. Recently, wheelchairs are widely used in daily lives by disabled person. In this paper, a new power assist control using acceleration information of object is proposed as intelligent function of wheelchair. In this strategy, acceleration information of object is utilized to generate the power assist torque for wheels driving. The proposed approach brings a sophisticated function that makes it easy to carry out pushing task by wheelchair users. First, an equivalent acceleration observer (EAOB) is employed to use acceleration of object. Relative distance between wheelchair and object is kept by assuming acceleration of object input to EAOB. Second, a compliance control is constructed to generate the motion command of pushing task. Here the motion command is determined so that the wheelchair accelerates. The compliance control itself is passive control, and stable and flexible responses are expected. Thirdly, a disturbance observer (DOB) is employed to guarantee stability of whole control system. Here an assistance ratio is designed based on velocity control of wheelchair. Finally, to confirm the validity of the proposed approach in experiments, two test operations are implemented.

Original languageEnglish
Title of host publicationProceedings - 34th Annual Conference of the IEEE Industrial Electronics Society, IECON 2008
PublisherIEEE Computer Society
Number of pages6
ISBN (Print)9781424417667
Publication statusPublished - 2008 Jan 1

Publication series

NameIECON Proceedings (Industrial Electronics Conference)

ASJC Scopus subject areas

  • Control and Systems Engineering
  • Electrical and Electronic Engineering


Dive into the research topics of 'Pushing task by repulsive compliance control in electric wheelchair using acceleration information'. Together they form a unique fingerprint.

Cite this