UAV Auto-Return Navigation Using Multi-Camera Feature Mapping
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Aerial vehicles using a single camera face challenges in auto-returning due to differences in images captured during different orientations, preventing effective visual cognition for navigation.
Innovation Solution
Employing multiple cameras with diverse fields of view on an aerial vehicle to collect comprehensive image data, processing and storing selected features like feature points for navigation, and using these features to guide the auto-return path.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If multiple cameras with different fields of view are used to collect comprehensive image data for auto-return navigation, then the reliability and accuracy of visual cognition is improved, but the processing and memory storage burden increases
Solution Approach 1:
The patent extracts only the essential feature points from images rather than processing entire image data. This extraction approach reduces the data volume significantly while preserving the navigational information needed for auto-return, thereby reducing processing and storage burden while maintaining navigation reliability.
Solution Approach 2:
The patent applies different processing strategies to different parts of the image data. Feature points that are critical for navigation are extracted and stored with higher priority, while other image data is processed or discarded. This local quality approach ensures that the most important navigational information is preserved while reducing overall data processing requirements.
2Measurement precision
If feature points are selected and stored based on their presence or absence in multiple cameras, then the navigational accuracy is improved, but the selection complexity and processing time increase
Solution Approach 1:
The patent performs preliminary action by pre-processing images to identify and extract feature points during the initial data collection phase. By preparing and organizing feature points in advance with metadata about their presence in different cameras, the system reduces the computational burden during real-time navigation, thereby reducing processing time while maintaining accuracy.
3Area of stationary object
If images are collected at regular intervals during flight, then the coverage of the flight path is improved, but the data storage requirements increase
Solution Approach 1:
The patent extracts only the essential feature points from images rather than storing entire images. This extraction approach reduces the data volume significantly while preserving the navigational information needed for auto-return, thereby reducing storage requirements while maintaining comprehensive flight path coverage.
Data Source
AI summary
An apparatus of controlling flight of an aerial vehicle includes a memory storing a computer program code, and one or more processors, individually or collectively, configured to execute the computer program code to: collect, while the aerial vehicle traverses a flight path, a set of images corresponding to different fields of view of an environment around the aerial vehicle, construct a map of the environment using the set of images, and control the aerial vehicle to return along a return path using the map of the environment.