Automated 3D Building Model Estimation for Solar Design
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing software solutions for optimizing solar panel installation require substantial user input, making the design process tedious and inefficient for estimating the usable surface area of roofs for solar energy systems, which complicates the determination of optimal solar panel placement and energy production.
Innovation Solution
An automated 3D building model estimation system using aerial imagery and 3D data, employing machine learning models to predict roof outlines, pitches, and heights, minimizing user input and requiring only a building address, utility rates, and energy bill data to generate a virtual solar energy system design.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If automated 3D building model estimation is implemented using machine learning models, then productivity is improved, but device complexity increases
Solution Approach 1:
The system segments the complex task of 3D building model estimation into distinct machine learning models: a first model predicts roof outlines from aerial imagery, a second model predicts pitches and heights from outlines and 3D data, and a third model determines usable surface area. This segmentation improves productivity by enabling automated processing while managing complexity through modular architecture.
Solution Approach 2:
The patent introduces intermediary processing steps between data input and final output: aerial imagery and 3D data are pre-processed to extract features, which then serve as inputs to sequential machine learning models. These intermediaries transform raw data into structured predictions, improving overall system efficiency while distributing computational complexity across multiple specialized components.
2Ease of operation
If minimal user input is required, then ease of operation is improved, but measurement precision may worsen
Solution Approach 1:
The system replaces manual measurement and user-driven data collection with automated machine learning models that process aerial imagery and 3D data. This substitution maintains high measurement precision by using trained models to extract roof geometry features automatically, while dramatically improving ease of operation by requiring only minimal user input such as building address and energy bill data.
Data Source
AI summary
Automated three-dimensional (3D) building model estimation is disclosed that predicts roof top outlines, pitches and heights based on imagery and 3D data. In an embodiment, a method comprises: obtaining an aerial image of a building based on an input address; obtaining three-dimensional (3D) data containing the building based on the input address; pre-processing the aerial image and 3D data; reconstructing a 3D building model from the pre-processed image and 3D data, the reconstructing including: predicting, using instance segmentation, a mask for each roof component of the building; predicting, using a first machine learning model with the mask as input, an outline for each roof component; predicting, using a second machine learning mode with the mask and outline as input, a pitch and height of each roof component; and rendering the 3D building model based on the predicted outline, pitch and height of each roof component.