UAV Imaging Fusion for Thermal-Aware Object Avoidance
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Unmanned aerial vehicles (UAVs) face challenges in accurately detecting and avoiding warm-blooded and non-warm-blooded objects during delivery, as existing systems rely solely on depth information, which can lead to false detection and safety risks due to the inability to differentiate between warm-blooded and non-warm-blooded objects based on protrusion alone.
Innovation Solution
The integration of depth and thermal information using multiple cameras, where depth maps are combined with thermograms to provide a 3D and thermal representation of the scene, allowing the UAV to differentiate between warm-blooded and non-warm-blooded objects by analyzing thermal signatures.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If depth information alone is used for object detection, then the system is simple and fast, but detection accuracy deteriorates due to inability to differentiate warm-blooded from non-warm-blooded objects
Solution Approach 1:
The patent combines depth information from a first camera and thermal information from a second camera into a unified detection system. The processor integrates both data types to generate comprehensive object detection, enabling differentiation between warm-blooded and non-warm-blooded objects while maintaining system efficiency through unified processing architecture
2Reliability
If multiple camera types are integrated for improved detection, then object differentiation capability is improved, but device complexity increases
Solution Approach 1:
The processor is designed to handle multiple types of imaging data (depth and thermal) through a single unified processing pipeline. This multi-functional approach enables the system to perform both depth-based object detection and thermal-based warm-blooded object identification using the same processing hardware, reducing overall system complexity while improving reliability
3Measurement precision
If thermal information is added to depth information, then false detection is reduced, but processing requirements increase
Solution Approach 1:
The system processes thermal and depth information selectively based on detection needs. The processor integrates thermal data primarily for warm-blooded object identification while using depth information for general object detection, avoiding full processing of all data types in all scenarios and thereby reducing unnecessary power consumption while maintaining high classification accuracy
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
This solution enhances the accuracy of object detection and avoidance, improving safety by reducing false positives and ensuring that deliveries are aborted or rerouted to avoid warm-blooded objects while allowing safe passage over non-warm-blooded objects, such as furniture or appliances.
Implementation Method 1
one or more cameras that are configured to obtain images of a scene using visible light that are converted into a depth map
Implementation Method 2
one or more other cameras that are configured to form images of the scene using infrared radiation (IR)
Implementation Method 3
a third camera that is approximately equidistant between the first camera and the second camera. The third camera is configured to form a third image, or thermogram, of the scene using infrared radiation
Data Source
AI summary
Described is an imaging component for use by an unmanned aerial vehicle (“UAV”) for object detection. As described, the imaging component includes one or more cameras that are configured to obtain images of a scene using visible light that are converted into a depth map (e.g., stereo image) and one or more other cameras that are configured to form images, or thermograms, of the scene using infrared radiation (“IR”). The depth information and thermal information are combined to form a representation of the scene based on both depth and thermal information.