FITTING A PARAMETRIC MODEL TO A CLOUD OF POINTS VIA OPTIMIZATION METHODS

<p>Computer Aided Design (CAD) is a powerful tool for designing</p> <p> parametric geometry. However, many CAD models of current</p> <p> configurations are constructed in previous generations of CAD</p> <p> systems, which represent the configuration simply as a collection of</p> <p> surfaces instead of as a parametrized solid model. But since many</p> <p> modern analysis techniques take advantage of a parametrization, one</p> <p> often has to re-engineer the configuration into a parametric</p> <p> model. The objective here is to generate an efficient, robust, and</p> <p> accurate method for fitting parametric models to a cloud of</p> <p> points. The process uses a gradient-based optimization technique,</p> <p> which is applied to the whole cloud, without the need to segment or</p> <p> classify the points in the cloud a priori.</p> <p> First, for the points associated with any component, a variant of</p> <p> the Levenberg-Marquardt gradient-based optimization method (ILM) is</p> <p> used to find the set of model parameters that minimizes the</p> <p> least-square errors between the model and the points. The</p> <p> efficiency of the ILM algorithm is greatly improved through the use</p> <p> of analytic geometric sensitivities and sparse matrix techniques.</p> <p> Second, for cases in which one does not know a priori the</p> <p> correspondences between points in the cloud and the geometry model's</p> <p> components, an efficient initialization and classification algorithm</p> <p> is introduced. While this technique works well once the</p> <p> configuration is close enough, it occasionally fails when the</p> <p> initial parametrized configuration is too far from the cloud of</p> <p> points. To circumvent this problem, the objective function is</p> <p> modified, which has yielded good results for all cases tested.</p> <p> This technique is applied to a series of increasingly complex</p> <p> configurations. The final configuration represents a full transport</p> <p> aircraft configuration, with a wing, fuselage, empennage, and</p> <p> engines. Although only applied to aerospace applications, the</p> <p> technique is general enough to be applicable in any domain for which</p> <p> basic parametrized models are available.</p>

Name thesis:degree_name

Doctor of Philosophy (PhD)

Level thesis:degree_level

Dissertation

Discipline thesis:degree_discipline

Mechanical and Aerospace Engineering

Year

2017

Author dc:creator

• Jia, Pengcheng

Contributors dc:contributor

• John F. Dannenhoffer
• Sinéad C. Mac Namara

• Engineering

Repository record dc:identifier

https://surface.syr.edu/etd/673

OAI identifier oai:identifier

oai:surface.syr.edu:etd-1673