UAV 3D Scan Planning With Iterative Model Refinement
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional UAV systems lack the capability for real-time, autonomous 3D scanning of complex structures with concavities, irregular surfaces, and oblique geometries, requiring manual operation and extensive human intervention for accurate data capture.
Innovation Solution
The implementation of an unmanned aerial vehicle (UAV) system that autonomously scans 3D targets by generating a lower-resolution model, iteratively refining it in real-time, and dynamically updating the scan plan to achieve high-resolution 3D reconstruction, using onboard processors and image sensors for navigation and data fusion.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If manual operation is used for UAV scanning, then human intervention can ensure accurate data capture, but the operation complexity and time consumption increase significantly
Solution Approach 1:
The UAV system performs autonomous scanning by automatically generating scan plans based on lower-resolution models, executing scans, and iteratively refining higher-resolution models without continuous human intervention. The system serves itself by making autonomous decisions about scan path planning, pose determination, and model refinement throughout the scanning process.
Solution Approach 2:
Before performing high-resolution scanning, the system first generates a lower-resolution 3D model of the target. This preliminary model is used to plan the scan path and determine optimal UAV poses, allowing the system to prepare and organize the scanning process in advance, thereby improving efficiency and reducing operational complexity.
2Measurement precision
If conventional scanning methods are used for complex structures with concavities and irregular surfaces, then complete coverage can be achieved, but the scanning time and number of re-scans increase
Solution Approach 1:
The scan plan is dynamically adjusted during the scanning process based on the iteratively refined higher-resolution 3D model. As the model accuracy improves, the system can adaptively modify the scan path and UAV poses to optimize coverage of complex geometries, thereby reducing redundant scans and completing the process more efficiently.
Solution Approach 2:
The system uses feedback from the lower-resolution model to plan scans and continuously refines the higher-resolution model based on captured images. This iterative feedback loop allows the system to identify areas requiring additional scanning and adjust the scan plan accordingly, ensuring complete coverage while minimizing unnecessary re-scanning.
3Measurement precision
If high-resolution scanning is performed directly without preliminary modeling, then final model accuracy is improved, but the computational load and processing time increase
Solution Approach 1:
The scanning process is segmented into two stages: first generating a lower-resolution model for scan planning, then performing high-resolution scanning. This segmentation divides the computational task into manageable parts, reducing the immediate computational load while still achieving high final model accuracy through iterative refinement.
Solution Approach 2:
The system performs preliminary scanning to create a lower-resolution model before executing the final high-resolution scan. This preliminary action provides a framework for planning the detailed scan, allowing the system to focus computational resources efficiently and reduce overall processing time while maintaining high accuracy.
4Extent of automation
If autonomous scanning is implemented for complex targets, then human intervention is reduced, but the system must handle complex geometries with concavities and irregular surfaces
Solution Approach 1:
The system transitions from 2D image capture to 3D model reconstruction by determining UAV poses based on normals to points in the lower-resolution model. This dimensional transition enables the autonomous system to handle complex 3D geometries including concavities and irregular surfaces by utilizing spatial information and surface orientation data.
Data Source
AI summary
In some examples, one or more processors of an unmanned aerial vehicle (UAV), control a propulsion mechanism of the UAV to cause the UAV to navigate to a plurality of positions in relation to a scan target. Using one or more image sensors of the UAV, a first image of the scan target is captured from a first position of the plurality of positions, and a second image of the scan target is captured from a second position of the plurality of positions. A disparity is determined between the first image captured at the first position and the second image captured at the second position. A three-dimensional model corresponding to the scan target is determined based in part on the disparity determined between the first image and the second image.