Radar calibration method and device and storage medium

Abstract

The invention discloses a radar calibration method and device and a storage medium, and the method comprises the steps: obtaining a two-dimensional image of a camera and a two-dimensional point cloud image of a radar at the same time; determining a line segment representing the detection target in the two-dimensional image, and obtaining a plurality of points corresponding to the line segment position in the two-dimensional point cloud image to form a point-line pair set; and determining a plurality of point-line pairs meeting a preset condition, constructing a plurality of point-line constraints, and randomly selecting a preset number of point-line constraints from the plurality of point-line constraints as target point-line constraints. After converting into linear constraint, solving a closed-form solution of the linear constraint and obtaining an initial solution of a conversion matrix; and performing nonlinear optimization on the initial solution of the conversion matrix by using the remaining point-line constraints to obtain an accurate conversion matrix for calibrating the radar. According to the technical scheme provided by the invention, the technical problems of high calibration time and labor cost caused by performing radar calibration through a calibration plate and poor model robustness easily caused by using a point-to-point constraint direct solution conversion matrix in the prior art can be solved.

Description

technical field [0001] The present invention relates to the technical field of radar, and in particular, to a radar calibration method, device and storage medium. Background technique [0002] Radar and camera are important sensors in vehicle systems, which play a key role in object recognition and localization. Both can be used for lane detection. Radar and camera have their own advantages and disadvantages. Among them, the camera has a strong ability to identify obstacles and can classify the obstacles in detail, but it is difficult for the camera to accurately locate the object, and the data collection is limited by the ambient light. Radar has a long detection distance and can accurately obtain three-dimensional information of objects, but its ability to identify obstacles is weak, and it is difficult to classify objects. [0003] Integrating the functions of radar and camera can complement each other and improve the performance of the radar system. Among them, the firs...

AI Technical Summary

Problems solved by technology

[0009] The present invention provides a radar calibration method, device and storage medium, aiming to effectively solve the problem of high calibration time and labor costs caused by radar calibration through calibration boards in the prior art, and poor robustness of the model due to point-to-point constrained solution of the Euclidean transformation matrix technical issues

Images