A method, system, and intelligent driving vehicle for calibrating the rotation angle of an onboard millimeter-wave radar.

By filtering and fitting millimeter-wave radar point cloud data, calculating the rotation angle and performing regression processing, the measurement distortion problem caused by rotation angle deviation was solved, improving the accuracy and stability of target recognition and autonomous driving systems.

CN122085231APending Publication Date: 2026-05-26DONGFENG MOTOR GRP
View PDF 0 Cites 0 Cited by

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
DONGFENG MOTOR GRP
Filing Date
2026-03-19
Publication Date
2026-05-26

AI Technical Summary

Technical Problem

In existing technologies, millimeter-wave radar suffers from problems in target recognition accuracy and autonomous driving decision-making and control layer response accuracy due to rotation angle deviation in practical use, especially causing height errors and lateral position calculation deviations in adjacent lane target detection scenarios.

Method used

By collecting raw point cloud sets during vehicle movement, filtering target point cloud sets, dividing point cloud subsets and performing linear fitting, calculating the included angle, and combining regression peak value processing, the rotation angle is accurately calibrated and rotation angle deviation is compensated.

Benefits of technology

It significantly improves the accuracy of target measurement and identification, provides reliable environmental perception data, and ensures the response accuracy and stability of the autonomous driving system.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN122085231A_ABST
    Figure CN122085231A_ABST
Patent Text Reader

Abstract

This invention discloses a method, system, and intelligent driving vehicle for calibrating the rotation angle of an onboard millimeter-wave radar. The method includes: when the onboard millimeter-wave radar is invoked for the first time, acquiring an original point cloud set within a set spatial range using the onboard millimeter-wave radar according to a set scanning cycle; filtering the original point cloud set to obtain a target point cloud set; dividing the target point cloud set into several point cloud subsets along the X-axis; for each point cloud subset, fitting a straight line based on the two-dimensional coordinates of all points in the subset, and calculating the angle between the straight line and the X-Y horizontal reference plane; performing regression peak value processing on the angles of all point cloud subsets to obtain the radar rotation angle corresponding to a set scanning cycle; performing secondary regression peak value processing on the radar rotation angles corresponding to all set scanning cycles to obtain the current calibration rotation angle of the onboard millimeter-wave radar; wherein, the current calibration rotation angle is used to compensate for the target detection data.
Need to check novelty before this filing date? Find Prior Art