UAV Backup Pose Planning for Obstructed 3D Scanning
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current unmanned aerial vehicle (UAV) systems lack the capability for efficient, autonomous scanning of complex structures with concavities, irregular surfaces, and oblique geometries, requiring manual operation and resulting in incomplete or inaccurate 3D reconstructions.
Innovation Solution
The implementation of a UAV system that autonomously scans 3D targets by generating a lower-resolution model, dynamically updating the scan plan, and improving resolution in real-time using onboard processors and image sensors, enabling consistent framing and robust feature detection through machine learning or human review.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If manual operation is used to scan complex structures, then the UAV can reach difficult vantage points, but the scanning efficiency and accuracy deteriorate due to human intervention requirements
Solution Approach 1:
The UAV system performs autonomous scanning operations by independently generating scan plans, navigating to capture positions, and reconstructing 3D models without continuous human intervention. The system serves itself by automatically processing images and updating scan plans in real-time during flight.
Solution Approach 2:
The system generates an initial scan plan before the UAV takes off, pre-calculating capture positions and trajectories. This preliminary planning enables the UAV to autonomously execute the scanning mission without real-time human control, improving scanning efficiency.
2Device complexity
If a single-pass scanning approach is used, then the scanning process is simple, but the reconstruction accuracy deteriorates for complex structures with concavities and irregular surfaces
Solution Approach 1:
The scan plan is dynamically updated during the UAV flight based on images captured in real-time. The system continuously generates new scan plans that adapt to the actual scanning progress and identified gaps, enabling multi-pass scanning without pre-defining all trajectories. This dynamic approach ensures complete coverage of complex structures while maintaining operational simplicity.
3Measurement precision
If high-resolution scanning is performed on large areas, then the reconstruction accuracy is improved, but the scanning time increases significantly
Solution Approach 1:
The scanning process is divided into multiple passes, with each pass capturing images at specific resolutions. The system segments the large area into regions that require different levels of detail, performing high-resolution scanning only on areas needing it while using lower resolution for broader coverage areas, thus reducing total scanning time.
Solution Approach 2:
The system performs scanning at varying resolutions based on the specific requirements of different areas. Not all regions are scanned at maximum resolution - the system applies partial high-resolution scanning only where necessary for accurate 3D reconstruction, balancing accuracy requirements with time constraints.
4Device complexity
If the scan plan is fixed before flight, then the planning process is simple, but the system cannot adapt to obstructions or changing conditions during scanning
Solution Approach 1:
The system uses feedback from images captured during flight to continuously update and refine the scan plan. The processed images provide information about the current scanning progress and identified gaps, which feed back into the planning system to generate updated trajectories and capture positions, enabling real-time adaptation to obstructions and changing conditions.
Solution Approach 2:
An initial scan plan is generated before flight as a preliminary action, providing a starting framework. However, this preliminary plan is quickly superseded by real-time updates based on actual scanning conditions, combining the benefits of pre-planning with adaptive responsiveness.
Data Source
AI summary
In some examples, an unmanned aerial vehicle (UAV) may access a scan plan that includes a sequence of poses for the UAV to assume to capture images of a scan target using one or more image sensors. The UAV may check a next pose of the scan plan for obstructions. Responsive to detection of an obstruction, the UAV may determine a backup pose based at least on a field of view of the next pose. The UAV may control a propulsion mechanism to cause the UAV to fly to assume the backup pose. The UAV may capture, based on the backup pose and using the one or more image sensors, one or more images of the scan target.