Sensorless Tool Collision Detection for Multi-axis Machine Tools by Integration of Disturbance Information

Tetsuya Shigematsu, Ryo Koike, Yasuhiro Kakinuma, Tojiro Aoyama, Kouhei Ohnishi

Research output: Contribution to journalConference articlepeer-review

5 Citations (Scopus)

Abstract

Tool collision must be detected with high response and reliability to reduce damage on machine-tool components. Several researches have proposed tool collision detection methods by monitoring a fluctuation in acceleration or force in collision direction with additional sensors. However, the installation of additional sensors is not desirable due to cost, machine-tool stiffness and failure rate. Moreover, these approaches do not focus on practical use of information in the other directions, although the collision-induced fluctuation can be observed not only in collision direction. This paper presents a tool collision detection method by applying disturbance observer theory and integrating the estimated collision force information in two axes. Furthermore, hotelling's T2 statistic is employed as a multivariate statistical process to the estimated collision force in the two directions. The proposed method successfully detects tool collision with higher response and reliability than that using only information in collision direction.

Original languageEnglish
Pages (from-to)658-663
Number of pages6
JournalProcedia CIRP
Volume57
DOIs
Publication statusPublished - 2016
Event49th CIRP Conference on Manufacturing Systems, CIRP-CMS 2016 - Stuttgart, Germany
Duration: 2016 May 252016 May 27

Keywords

  • breakage
  • disturbance observer
  • in-process monitoring
  • sensorless
  • tool collision

ASJC Scopus subject areas

  • Control and Systems Engineering
  • Industrial and Manufacturing Engineering

Fingerprint

Dive into the research topics of 'Sensorless Tool Collision Detection for Multi-axis Machine Tools by Integration of Disturbance Information'. Together they form a unique fingerprint.

Cite this