UAV Monocular Distance Estimation Without Level-Ground Assumptions
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for estimating the distance to a target by unmanned aerial vehicles (UAVs) using monocular cameras are inaccurate due to unstable height data and rely on the assumption of a level ground, and are unable to estimate distance when the UAV is hovering.
Innovation Solution
A method that acquires image frames, location information including height and pixel coordinates, and attitude information such as pitch angle to calculate the distance to the target using geometric calculations involving angles and focal lengths, allowing for real-time accurate estimation of spatial location without assuming a level ground.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If distance is calculated using height data from the unmanned aerial vehicle, then the calculation can be performed, but the accuracy deteriorates due to unstable height data and deviation accumulation
Solution Approach 1:
The patent introduces an intermediary geometric calculation method that uses the photographing device's pitch angle and target's pixel coordinates as intermediate parameters to derive distance, rather than directly using unstable height data. This intermediary approach filters out the instability in height measurements.
Solution Approach 2:
The patent changes the parameters used for distance calculation from direct height data to a combination of pitch angle and pixel coordinates. By transforming the calculation parameters, the system achieves more stable and accurate distance measurement that is not directly affected by height data instability.
2Measurement precision
If triangulation method is used to estimate target location, then distance can be estimated, but the method fails when the unmanned aerial vehicle is hovering without displacement
Solution Approach 1:
The patent transforms the static triangulation method into a dynamic solution by incorporating the photographing device's pitch angle information. This allows the system to calculate distance based on the relationship between pitch angle changes and target pixel position, making it applicable whether the UAV is hovering or moving.
Solution Approach 2:
The patent creates a universal distance calculation method that works across multiple flight conditions (hovering, moving forward, moving backward) by using parameters (pitch angle and pixel coordinates) that are available in all conditions, rather than relying on displacement which is zero during hovering.
3Productivity
If traditional methods are used for target detection and following control, then the system can operate, but computational complexity increases and real-time performance deteriorates
Solution Approach 1:
The patent extracts only the essential parameters needed for distance calculation (pitch angle and pixel coordinates) from the complex image data, discarding unnecessary information. This extraction approach simplifies computation while maintaining accuracy, enabling real-time processing.
Data Source
AI summary
The present invention relates to a method and device for estimating a distance to a target, and a unmanned aerial vehicle. The method is applicable to a unmanned aerial vehicle including a photographing device. The method includes: acquiring a current frame of image of the target captured by the photographing device; acquiring location information of the target according to the current frame of image, where the location information includes a height of the target and two-dimensional pixel coordinates of the target in the image; acquiring attitude information of the photographing device, where the attitude information includes a pitch angle of the photographing device; and acquiring a distance between the target and the unmanned aerial vehicle according to the location information of the target and the attitude information of the photographing device.