3D Simulation Environment Generation Using Geographic Land Use Sectors
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Creating realistic three-dimensional simulation environments for autonomous and semi-autonomous driving tests is time-consuming due to the need for manual generation of complex 3D scenes, often with insufficient data, requiring extensive manual work and human creativity.
Innovation Solution
A computer-implemented method that uses a basic library of three-dimensional virtual objects, captures geographic region characteristics, derives additional information, and populates sectors with matching objects based on land use and climate data, allowing for automatic creation of realistic 3D environments, even with limited data availability.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Manufacturing precision
If manual generation of complex 3D scenes is used, then realistic simulation environments can be created, but the process becomes very time-consuming and labor-intensive
Solution Approach 1:
The patent uses satellite images and aerial photographs as templates to automatically generate 3D scenes. Real-world geographic data is copied and transformed into virtual 3D representations, eliminating the need for manual modeling while preserving realistic geographic features and land use patterns.
Solution Approach 2:
The patent replaces manual mechanical modeling processes with automated computer-based image processing and 3D reconstruction algorithms. Software automatically generates 3D terrain models, buildings, and vegetation from 2D satellite imagery, substituting human creative work with computational processes.
2Manufacturing precision
If comprehensive data is collected for accurate 3D scene generation, then realism improves, but data acquisition becomes more difficult and time-consuming
Solution Approach 1:
The patent uses satellite images and aerial photographs that serve multiple purposes simultaneously: they provide geographic layout, land use information, terrain elevation data, and spatial relationships all in one data source. This multi-functional data reduces the need to collect separate datasets for different geographic attributes.
Solution Approach 2:
Existing satellite imagery and aerial photographs that already contain comprehensive geographic information are copied and processed. This approach leverages pre-existing data that has already been captured by remote sensing systems, eliminating the need for time-consuming field surveys and manual data collection.
3Reliability
If random testing with virtual vehicles is performed, then comprehensive test coverage is achieved, but the testing process becomes inefficient without structured organization
Solution Approach 1:
The patent divides the geographic region into multiple sectors based on land use types (residential, commercial, industrial, agricultural, natural areas). This segmentation allows test scenarios to be systematically organized by geographic zone, enabling efficient coverage of different environmental conditions while maintaining structured test management.
Data Source
Figure 1
Figure 2
AI summary
The invention relates to a computer-implemented method for creating a three-dimensional simulation environment, comprising the following method steps: providing a basic library of three-dimensional virtual objects, capturing the input of a geographical region, obtaining characteristic information that characterizes features of different areas of the geographical region, deriving additional information for the different areas of the geographical region based on the characteristic information, wherein land use information is derived as additional information if the characteristic information does not include land use information, identifying the objects from the basic library that occur in the geographical region based on the characteristic information and/or the additional information, and storing these objects in a regional library.Dividing the geographic region into sectors of similar land use based on land use information, identifying objects from the regional library that match the land use of each sector, and populating each sector with a selection of objects identified for that sector based on its land use. This provides a way to create a realistic virtual three-dimensional environment in a simple and reliable manner.