AI-Generated TIN Layered Data Structure for 3D Design
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The existing techniques for utilizing AI-generated triangulated irregular networks (TINs) in civil engineering are inefficient, as they require manual editing of a large number of triangles in a 3D design model, making it tedious and counterintuitive, and hinder iterative refinement, as users must recreate the model from scratch after each optimization.
Innovation Solution
A layered data structure approach is introduced, defining site objects through multiple phases: conceptual, preliminary, and final, with data structures that allow for efficient propagation of changes, reducing the need for manual editing and enabling seamless integration of AI-generated TINs into 3D design models.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Extent of automation
If the number of triangles in the TIN is increased to enable sufficient AI optimization, then the optimization capability is improved, but the ease of manual editing and operation deteriorates
Solution Approach 1:
The patent segments the TIN into multiple hierarchical levels or groups, allowing users to edit elevations at a higher level rather than individually editing each triangle. This segmentation enables AI optimization to work with detailed triangle data while users interact with simplified grouped representations, resolving the contradiction between automation capability and ease of operation.
Solution Approach 2:
The patent introduces an intermediary layer between the AI optimization process and user interaction. This intermediary manages the TIN data structure, allowing AI to optimize individual triangles while presenting a simplified interface to users for bulk elevation adjustments. The intermediary translates between detailed optimization needs and simplified user operations.
2Adaptability or versatility
If the TIN is used directly in detailed 3D design model, then the integration is improved, but the productivity deteriorates due to tedious manual editing
Solution Approach 1:
The patent performs preliminary organization and grouping of TIN triangles before the user needs to edit them. By pre-segmenting the TIN into manageable groups or levels during the AI optimization phase, the system prepares the data structure in advance to enable efficient bulk editing operations later, avoiding the need for tedious individual triangle editing during detailed design modeling.
3Manufacturing precision
If the traditional workflow of recreating 3D design model from scratch is used after optimization, then the model accuracy is improved, but the loss of time increases
Solution Approach 1:
The patent enables the AI-optimized TIN to be copied or transferred directly into the detailed 3D design model rather than recreating the model from scratch. The optimized elevation data is preserved and reused across different design phases, maintaining model accuracy while eliminating the time-consuming recreation process. This copying approach allows iterative refinement without starting over.
Data Source
AI summary
In example embodiments, techniques are provided for enabling use of an AI-generated TIN in generation of a 3D design model by defining site objects (e.g., pads) using multiple (e.g., three) phases (i.e. states). A conceptual phase may be associated with a conceptual data structure, a preliminary phase may be associated with the conceptual data structure and a preliminary data structure, a final phase may be associated with the conceptual data structure, the preliminary data structure, and a final data structure. If changes are made in the conceptual phase, for example, as a result of AI optimization, they may be propagated up to the preliminary data structure and final data structure via the vertical draping. Changes made in the preliminary phase or final phase may be propagated down to the conceptual data structure by treating boundaries and breaklines as spatial constraints.