A potential field based multi-axis printing path generation algorithm

  • Fubao Xie
  • , Danjie Bi
  • , Kai Tang*
  • *Corresponding author for this work

Research output: Contribution to journalArticlepeer-review

Abstract

Robotic multi-axis curved layer additive manufacturing has drawn more attention recently owing to its unique ability of printing complex shapes while avoiding the stair-step effect and the usage of support structure. However, as the robotic manipulator is an open serial structure and the printing layer now is no longer planar but a free-form surface, the shape and pattern of a printing path now have a huge impact on the mechanical behaviour of the printer–an arbitrarily planned printing path could cause severe angular fluctuations of the joints of the robotic manipulator which in turn would lead to poor printing quality and prolonged printing time. In this paper, a potential field based algorithm is proposed for automatically generating a printing path for printing an arbitrary free-form curved layer on a robotic multi-axis printer. By considering both the printing efficiency and the joints’ kinematics, the potential field identifies the local optimal feed direction on the curved layer. Then, by employing the iso-cusp height expansion method, a printing path is generated to best fit this potential field. Both computer simulations and physical printing experiments are carried out to assess the proposed methodologies and the results give a positive confirmation on their advantages.

Original languageEnglish
Pages (from-to)1277-1299
Number of pages23
JournalInternational Journal of Computer Integrated Manufacturing
Volume33
Issue number12
DOIs
StatePublished - 2020
Externally publishedYes

Keywords

  • Curved layer fused deposition modelling
  • multi-axis printing path generation
  • robotic 3D printing
  • robotic local-printability potential field

Fingerprint

Dive into the research topics of 'A potential field based multi-axis printing path generation algorithm'. Together they form a unique fingerprint.

Cite this