UAV Radar-Visual Joint Calibration via Pose Adaptation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Traditional data fusion methods for unmanned aerial vehicles (UAVs) fail to accurately integrate millimeter-wave radar and visual image data due to the lack of consideration for height-related information and pitch angle changes, resulting in significant deviations in depth information and poor data conversion.
Innovation Solution
A joint calibration method and apparatus that obtain and update calibration parameters based on the ground height and pitch angle of the detection radar, establishing a spatial conversion relationship between the radar and image acquisition devices, enabling accurate data fusion by using a three-dimensional data entry model and coordinate conversion functions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional data fusion methods are used without considering height and pitch angle, then the system complexity is low, but the measurement precision and reliability of depth information deteriorate significantly
Solution Approach 1:
The patent performs preliminary calibration by pre-establishing the spatial conversion relationship between radar and image acquisition devices at multiple known pose positions. This pre-calibration data is stored and later used to quickly determine the appropriate conversion relationship based on current pose information, avoiding the need for complex real-time calculations while maintaining high measurement precision
Solution Approach 2:
The patent introduces dynamic pose information (ground height and pitch angle) into the calibration system. By making the calibration parameters adaptive to changing pose conditions rather than fixed, the system maintains measurement precision across varying flight conditions without requiring a completely complex recalibration system
2Adaptability or versatility
If fixed calibration parameters are used without updating, then the ease of operation is high, but the adaptability to attitude changes deteriorates, causing calibration accuracy to decrease
Solution Approach 1:
The patent extends the calibration parameter space by adding pose information dimensions (ground height and pitch angle) to the traditional fixed calibration parameters. This creates a multi-dimensional calibration system where parameters are selected based on the current pose state, enabling adaptability to attitude changes while maintaining operational simplicity through automated selection
Solution Approach 2:
The patent implements a feedback mechanism where current pose information (ground height and pitch angle) is continuously obtained and used to select or update the appropriate calibration parameters. This closed-loop approach ensures the system automatically adapts to attitude changes without requiring manual intervention, balancing adaptability with ease of operation
Data Source
AI summary
Joint calibration methods implemented by an unmanned aerial vehicle and a server are disclosed. The server receives pose information of a detection radar from the unmanned aerial vehicle. The pose information includes a ground height and a pitch angle of the detection radar. The server determines target calibration parameters matching the pose information of the detection radar, and sends the target calibration parameters to the unmanned aerial vehicle. The unmanned aerial vehicle determines a spatial conversion relationship between the detection radar and an image acquisition device based on the target calibration parameters, and performs data fusion between the detection radar and the image acquisition device according to the spatial conversion relationship.