Radar-LiDAR Relative Pose Calibration Using a Target Obstacle
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Millimeter-wave radar and laser radar sensors in ADAS systems operate in different coordinate systems, necessitating a conversion to utilize detection information effectively, but existing methods are inadequate for determining the relative pose between these sensors accurately.
Innovation Solution
A method and apparatus that utilize a target obstacle detectable by both sensors to determine the relative pose by collecting and processing millimeter-wave data and sensor-specific data, employing a millimeter-wave radar signal booster to distinguish target obstacles and filter noise, and minimizing a target function to establish the relative pose between the sensors.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If coordinate system conversion is performed between millimeter-wave radar and laser radar sensors, then detection information can be utilized more effectively, but the accuracy of relative pose determination deteriorates due to insufficient existing methods
Solution Approach 1:
The patent introduces a calibration board with specific geometric features (corners, edges, circular arcs) as an intermediary object that both millimeter-wave radar and laser radar can detect. This calibration board serves as a common reference framework, enabling accurate relative pose determination between the two sensors with different coordinate systems while maintaining effective detection information utilization
2Ease of manufacture
If sensor calibration is performed using conventional methods, then the calibration process can be completed, but the calibration accuracy deteriorates due to inadequate handling of different sensor coordinate systems
Solution Approach 1:
The patent transforms the calibration problem by changing the parameter representation method. It uses geometric parameters of the calibration board (corner coordinates, edge equations, circular arc parameters) as intermediate parameters to establish the transformation relationship between coordinate systems, thereby improving calibration accuracy while maintaining process completeness
Solution Approach 2:
The patent replaces traditional mechanical alignment methods with a mathematical model-based approach. By using geometric feature detection and coordinate transformation algorithms, the system achieves more accurate calibration without relying on physical alignment procedures
Data Source
Figure 1~2
Figure 3
Figure 4
AI summary
Embodiments of the present disclosure discloses a method and an apparatus for determining a relative pose, a device, and a medium. The method comprises: obtaining millimeter-wave data for obstacles in a scene collected by a millimeter-wave radar sensor in a vehicle and first data for the obstacles in the scene collected by a first sensor in the vehicle, wherein, at least one millimeter-wave radar signal booster is disposed in the scene as a target obstacle and the obstacles comprise the target obstacle; determining first obstacle data for the target obstacle in a millimeter-wave radar coordinate system and second obstacle data for the target obstacle in a first sensor coordinate system according to the millimeter-wave data and the first data for the obstacles in the scene; and determining the relative pose between the millimeter-wave radar sensor and the first sensor according to the first obstacle data and the second obstacle data.