Geometric Reasoning

A better Technique to reduce user subjectivity

 

Geometry

General Overview

 

What is Geometric Reasoning for?

 

Geometric Reasoning is the main technique used in extracting information from 3D CAD models required for use in DFA analysis. The aim of the technique is to analyse the CAD model and generate information regarding component shape and complexity. This includes simple analyses such as the calculation of mass properties and more complex issues such as the study and identification of symmetry and major and minor axes. This information can then be used to validate assembly sequence and support the DFA analysis to reduce the workload of the design engineer when evaluating a design. Previous work has led to the development of an algorithm for the detection of symmetry from a CAD model. Future work will further develop this algorithm to identify additional characteristics that help Proactive DFA evaluate practical assembly issues such as component orientation, insertion trajectories and mating of components during assembly sequences.

 

Elimination of subjectivity


The role of geometric reasoning in the Designers’ Sandpit is to reduce subjective and time consuming user input by the automatic extraction of data already available within the CAD model of a product design. In particular, the validation and evaluation of the assembly sequence is geometry-dependent, as are many elements of the DFA methodology. However, the Designer’s Sandpit also addresses issues of concept design generation and therefore the links between the geometry of a product and its functionality will also be explored. In support of DFA analyses, geometric reasoning algorithms will be developed for the following purposes:

 

 

In support of assembly sequence validation and evaluation the following geometric analyses will be given priority:

 

  • Collision detection along component insertion trajectory
  • Analysis of stability at all stages of product assembly

 

Both these techniques require the identification of mating faces and the degrees of freedom at all component interfaces. This requires that the CAD system is configured in such a way to facilitate these interrogations and thus another major aspect of the geometric reasoning work package is to enable the input of this assembly data in a useful and unobtrusive manner. Finally, to support the concept modelling aspects of the system, the possibility of a constraint-based modelling environment that enables validation of product function with respect to component geometry, will be investigated.