Voxelization Method for Free Space Extraction in CAE

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Conventional methods for extracting free space from three-dimensional shape data in Computer Aided Engineering (CAE) impose a significant processing load due to the large number of finely divided face elements in complex products like automobiles, making outsourcing and data sharing inefficient.

Innovation Solution

A voxelization method that converts three-dimensional shape data into voxel structures, extends lines from cube faces to identify external regions, and performs inverse voxelization to isolate free space, reducing computational load and simplifying data representation.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If conventional methods are used to extract free space from three-dimensional shape data, then the free space can be identified, but the processing load increases significantly due to the large number of finely divided face elements

Engineering Contradiction:
Improvefree space extraction accuracyVSAvoidprocessing speed
Core Design Contradiction:
Measurement precisionVSProductivity

Solution Approach 1:

The patent segments the complex three-dimensional shape data into a simplified voxel grid structure. By dividing the space into discrete volumetric pixels (voxels) and representing the product shape as occupied voxels, the method transforms complex geometric data into a manageable binary representation, significantly reducing processing complexity while maintaining extraction accuracy

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent creates a simplified copy of the original three-dimensional shape data by generating a voxel model that replicates the essential geometric features. This voxel copy retains the necessary spatial information for free space extraction while eliminating the computational burden of the original finely divided face elements

Inventive Principle:
Principle #26Copying

2Measurement precision

If conventional methods extract free space data from complex three-dimensional shape data, then the free space shape is obtained, but the data complexity and processing requirements increase

Engineering Contradiction:
Improvefree space shape accuracyVSAvoiddata structure complexity
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The patent applies segmentation by converting continuous three-dimensional shape data into discrete voxel units arranged in a regular grid. This segmentation transforms complex geometric representations into a structured array of occupied and unoccupied voxels, simplifying the data structure while preserving the free space geometry for accurate extraction

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent changes the parameter representation from detailed face element geometry to a simplified voxel occupancy state (occupied/unoccupied). This parameter transformation reduces data complexity by replacing complex geometric descriptions with binary voxel states, making the data structure more manageable while maintaining the essential spatial information

Inventive Principle:
Principle #35Parameter changes

Data Source

PatentEP3282424B1Determination of free space in a product
Publication Date: 2019.12.18 FUJITSU LTD
  • EP3282424B1 patent drawingFigure 1
  • EP3282424B1 patent drawingFigure 2
  • EP3282424B1 patent drawingFigure 3

AI summary

A voxelization method includes: voxelizing a three-dimensional shape (300) to generate a first voxel structure (301) corresponding to the three-dimensional shape (300); specifying, in a case where lines perpendicular to respective faces of a cube or a cuboid (301a) containing the generated first voxel structure (301) are extended from the respective faces toward inside the cube or the cuboid (301a) until the lines hit the first voxel structure (301), a region (302) outside an outer periphery of the first voxel structure (301) according to whether at least lines extended from three faces orthogonal to each other intersect; setting the specified outside region (302) as a second voxel structure; and performing inversion to invert a region (303) of the voxel structures (301, 302) and a region not set as a voxel in the cube or the cuboid (301a).