Serialized Octree for Additive Manufacturing Layer Processing

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current additive manufacturing techniques face challenges in efficiently processing and retrieving data for layer-by-layer object generation, leading to inconsistent layer processing times and complex data retrieval processes.

Innovation Solution

The use of a serialized octree representation, ordered according to specific rules (Z-first ordering, parent nodes before descendants, and location encoding), allows for efficient data retrieval and processing, enabling consistent layer processing times by storing data in a structured format that facilitates rapid access.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If traditional data structures are used for additive manufacturing, then data can be stored, but data retrieval is complex and layer processing times are inconsistent

Engineering Contradiction:
Improvelayer processing time consistencyVSAvoiddata retrieval complexity
Core Design Contradiction:
ProductivityVSDevice complexity

Solution Approach 1:

The data structure is segmented into an octree hierarchy where the volumetric dataset is divided into octants at multiple levels. Each node represents a specific volumetric region and contains only the data necessary for that region, enabling efficient retrieval by accessing only relevant segments for each layer rather than processing entire datasets.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent transforms the traditional two-dimensional layer-wise data representation into a three-dimensional octree structure with explicit spatial hierarchy. This dimensional transformation allows data to be organized by spatial location and depth, enabling efficient navigation to specific layers through the 3D structure rather than sequential 2D array access.

Inventive Principle:
Principle #17Another dimension (Dimensionality change)

2Manufacturing precision

If detailed volumetric data is stored for high precision manufacturing, then manufacturing precision is improved, but memory access delays increase

Engineering Contradiction:
Improvevolumetric data precisionVSAvoidmemory access delay
Core Design Contradiction:
Manufacturing precisionVSLoss of time

Solution Approach 1:

The octree structure implements local quality by storing high-resolution volumetric data only in regions where precision is required, while using coarser representations in less critical areas. Each node in the octree can have different levels of detail based on the complexity of the geometry in that region, optimizing the balance between precision and memory access efficiency.

Inventive Principle:
Principle #3Local quality

Solution Approach 2:

The data is pre-organized into an octree structure before manufacturing begins, with all spatial relationships and hierarchical levels established in advance. This preliminary organization allows the system to quickly navigate to and retrieve only the specific volumetric data needed for each layer, avoiding the need to search through entire datasets during the manufacturing process.

Inventive Principle:
Principle #10Preliminary action

3Productivity

If data is organized for sequential layer processing, then manufacturing can proceed layer-by-layer, but concurrent processing of subsequent layers is not possible

Engineering Contradiction:
Improveconcurrent layer processing capabilityVSAvoiddata organization simplicity
Core Design Contradiction:
ProductivityVSEase of manufacture

Solution Approach 1:

The octree structure segments the volumetric data into independent spatial regions that can be processed in any order. Each node represents a self-contained volumetric region with all necessary information, allowing different layers and regions to be processed concurrently without requiring sequential dependencies, thus enabling parallel manufacturing operations.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

By organizing data in a 3D octree structure rather than sequential 2D layers, the patent enables random access to any volumetric region regardless of layer position. This dimensional reorganization allows the manufacturing system to process multiple layers concurrently by accessing different parts of the octree structure simultaneously, breaking the sequential processing constraint.

Inventive Principle:
Principle #17Another dimension (Dimensionality change)

Data Source

PatentUS10853998B2Serialising a representation of a three dimensional object
Publication Date: 2020.12.01 PERIDOT PRINT LLC
  • US10853998B2 patent drawing
  • US10853998B2 patent drawing
  • US10853998B2 patent drawing

AI summary

In an example, a method includes receiving, at a processor, data representing at least part of an object to be manufactured in a layer-by-layer manufacturing process. A serialised octree representation of at least part of the object may be generated from the data. In the serialised octree representation, nodes are ordered such that (i) a node representing a volume which includes a layer of the object to be generated earlier in an intended order of object generation precedes a node representing a volume which consists of layer(s) of the object to be generated subsequently; (ii) nodes representing a volume which includes a given layer of the object are ordered based on a level of the nodes within the octree representation, wherein parent nodes appear before descendent nodes; and (iii) nodes representing a volume which includes a given layer of the object and being of the same level within the octree representation are ordered according to a location encoding pattern.