An obstacle detection method and device, and related products

CN117169892BActive Publication Date: 2026-05-26BEIJING ZHIXINGZHE TECH CO LTD
View PDF 2 Cites 0 Cited by

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
BEIJING ZHIXINGZHE TECH CO LTD
Filing Date
2022-05-27
Publication Date
2026-05-26

AI Technical Summary

Technical Problem

Existing ultrasonic radars cannot achieve high-precision obstacle detection in harsh environments, and increasing the number of probes will increase costs and system complexity.

Method used

By generating and analyzing point cloud data, obstacle grids are determined, improving the obstacle perception accuracy of ultrasonic radar. The observation angle of the point cloud data is used to determine the obstacle grids.

Benefits of technology

Without increasing costs, it significantly improves the obstacle perception accuracy of ultrasonic radar, enhancing the adaptability of autonomous vehicles in rain, snow, and fog environments.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN117169892B_ABST
    Figure CN117169892B_ABST
Patent Text Reader

Abstract

This invention relates to an obstacle detection method and apparatus, and related products, comprising: acquiring detection distance data of each ultrasonic radar probe; determining the processing area of ​​each probe based on the detection distance data and detection range of each probe; segmenting the processing area of ​​the probe according to a preset interval to obtain multiple point cloud data corresponding to the probe; performing coordinate transformation on the relative coordinates of the point cloud data corresponding to the probe based on the installation angle of the probe on the vehicle and the current absolute positioning coordinates of the vehicle to obtain the global coordinates of the point cloud data in the absolute coordinate system; determining the grid in the grid map where the point cloud data is located based on the global coordinates of the point cloud data, and determining the angle interval of the angle distribution of the point cloud data; the grid has multiple preset angle intervals; counting the number of angle intervals of the angle distribution of all point cloud data contained in the grid within a cumulative time period, and determining whether the grid is an obstacle grid based on the number of angle intervals.
Need to check novelty before this filing date? Find Prior Art