Design-to-fabrication automation for the cognitive machine shop

  • Authors:

  • Kristina Shea;Christoph Ertelt;Thomas Gmeiner;Farhad Ameri

  • Affiliations:

  • Virtual Product Development Group, Institute of Product Development, Technische Universität München, Boltzmannstr. 15, 85748 Garching, Germany;Virtual Product Development Group, Institute of Product Development, Technische Universität München, Boltzmannstr. 15, 85748 Garching, Germany;Virtual Product Development Group, Institute of Product Development, Technische Universität München, Boltzmannstr. 15, 85748 Garching, Germany;Department of Engineering Technology, Texas State University, 601 University Dr., San Marcos, TX 78666, USA

  • Venue:
  • Advanced Engineering Informatics
  • Year:
  • 2010

Quantified Score

Hi-index 0.00

Visualization

Abstract

To meet the rising demands for pure customization of products, new approaches for automated fabrication of customized part geometry are needed, on both the software and hardware side, that balance flexibility, robustness and efficiency. This is a great challenge since today it requires significant human expertise supported, only partially, by computer-aided approaches. This paper introduces a new approach and framework for an autonomous design-to-fabrication system that integrates cognitive capabilities, such as reasoning from knowledge models and autonomous planning, and embeds these in the machines themselves to automatically fabricate customized parts. The framework integrates into a common process automatic workpiece selection using an ontology, generative CNC machining planning using shape grammars and automated fixture design, based on a novel flexible fixture device hardware. Initial results are given for the machining planning approach applied to 2.5D parts with a defined approach direction and the prototyped fixture device is presented. The advantages and potential of the framework stem mainly from applying the principles of cognitive technical systems to a fabrication system to develop an integrated and on-line approach. The methods are developed specifically for use on the machine shop floor to take advantage of the possibility to update and extend knowledge models to reflect current fabrication capabilities and to adapt to changes in the environment and re-plan during operation. Finally, future directions, including integrating on-line perception and learning, are discussed, which are required to create a truly flexible and cognitive fabrication system.