UAV Multi-Camera Auto-Return Using Feature Point Navigation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Aerial vehicles equipped with a single camera face challenges in auto-returning due to differences in images recorded during flight and return paths, leading to potential navigation errors without reliable external communication systems.
Innovation Solution
Equipping unmanned aerial vehicles (UAVs) with multiple cameras providing different fields of view to collect and compare image data, using feature points from images to navigate back to the original path, and selectively storing features based on parameters like movement and overlap to reduce processing demands.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If a single camera is used for aerial photography, then the device complexity is reduced, but the reliability of auto-return navigation deteriorates due to image segmentation and inability to recognize original path
Solution Approach 1:
The patent divides the imaging task into multiple segments by using multiple cameras with different fields of view (forward-facing, backward-facing, and side-facing cameras). Each camera captures images from its specific direction, and the system integrates these segmented views to form a complete environmental map, enabling reliable auto-return navigation without relying on external communication systems.
Solution Approach 2:
The patent makes the camera system multi-functional by equipping the aerial vehicle with multiple cameras that can serve different purposes: forward-facing cameras capture the original path during outward flight, backward-facing cameras capture the return path, and side-facing cameras provide lateral coverage. This universal imaging system enables both navigation and environmental mapping functions.
2Reliability
If multiple cameras with different fields of view are installed, then the reliability of auto-return navigation is improved, but the device complexity increases
Solution Approach 1:
The patent merges the functions of multiple cameras into a unified navigation system. The images from forward-facing, backward-facing, and side-facing cameras are integrated to create a comprehensive environmental map. This merging approach allows the system to achieve reliable auto-return navigation while managing the complexity through coordinated processing of multiple image sources.
Solution Approach 2:
The patent adds dimensional coverage by incorporating cameras facing different directions (forward, backward, sides) rather than relying on a single viewpoint. This multi-dimensional imaging approach ensures that the aerial vehicle can recognize its position and original path from multiple angular perspectives, improving navigation reliability.
3Measurement precision
If images from multiple cameras are collected and processed in real-time, then the navigation accuracy is improved, but the processing demands and energy consumption increase
Solution Approach 1:
The patent performs preliminary actions by collecting and storing images from multiple cameras during the outward flight path before the return journey. These pre-captured images serve as reference data for navigation during the return trip, allowing the system to compare current visual input with stored references without requiring intensive real-time processing of all images simultaneously, thus reducing energy consumption during critical navigation phases.
Data Source
AI summary
A method of controlling flight of an unmanned aerial vehicle (UAV) includes collecting, while the UAV traverses a flight path, a set of images corresponding to different fields of view of an environment around the UAV using multiple image capture devices. Each of the multiple image capture devices includes one of the different fields of view. The method further includes extracting a set of feature points from the set of images and selecting a subset of feature points from the set of feature points to be stored based on a number of image capture devices that collect images including the subset of feature points to aid navigation of the UAV.