Unlock instant, AI-driven research and patent intelligence for your innovation.

A long-distance obstacle detection method based on lidar multi-frame point cloud fusion

An obstacle detection and lidar technology is applied in the field of long-distance obstacle detection based on lidar multi-frame point cloud fusion, which can solve the problems of inability to effectively detect long-distance obstacles and sparse point clouds.

Active Publication Date: 2020-08-14
ZHEJIANG UNIV
View PDF6 Cites 1 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0005] The method of the present invention can solve the problem of being unable to effectively detect long-distance obstacles caused by the sparseness of single-frame laser laser point clouds

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • A long-distance obstacle detection method based on lidar multi-frame point cloud fusion
  • A long-distance obstacle detection method based on lidar multi-frame point cloud fusion
  • A long-distance obstacle detection method based on lidar multi-frame point cloud fusion

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0148] The present invention will be further described below in conjunction with accompanying drawings and examples.

[0149] The present invention uses the pictures provided by the public data set KITTI and the laser laser point cloud data to illustrate the implementation steps and effects of obstacle detection.

[0150] Step 1: Implement 1.1) of the present invention on the single-frame original point cloud, calculate the curvature of each point and extract feature points, where N 1 Value 5, Ths curv The value is 0.1. The obtained feature points are as figure 1 As shown, the smaller one is the original point cloud, and the larger one is the feature point.

[0151] Step 2: Implement 1.2) of the present invention on the single-frame original point cloud, obtain the inter-frame pose through inter-frame feature point matching, and remove motion distortion from the original point cloud to obtain the first de-distortion under the frame tail coordinate system point cloud. The ...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

PUM

No PUM Login to View More

Abstract

The invention discloses a long-distance obstacle detection method based on lidar multi-frame point cloud fusion. Establish the local coordinate system and the world coordinate system, calculate the extracted feature points of each laser point on the circular scanning line of the laser radar according to the original point cloud data in the local coordinate system, and obtain the current position through the matching of feature points between frames and map feature points Relative to the global pose of the starting position and the de-distorted point cloud in the world coordinate system; the de-distorted point cloud of the current frame and the previous frame is fused to obtain more dense de-distorted point cloud data and unified into the local coordinate system, and then to Two-dimensional grids are used for projection, and obstacles are screened out according to the height change characteristics of each two-dimensional grid. The invention solves the problem of low detection rate of distant obstacles caused by sparse laser point clouds, can effectively detect distant obstacles, has low false detection rate and missed detection rate, and can greatly reduce system cost.

Description

technical field [0001] The invention relates to a robot obstacle detection method, in particular to a long-distance obstacle detection method based on lidar multi-frame point cloud fusion for unmanned vehicle navigation. Background technique [0002] In the field of autonomous navigation of unmanned vehicles, the traditional methods for obstacle detection include detection methods based on stereo vision, structured light, millimeter-wave radar and lidar. The detection method based on stereo vision uses the color, edge and texture features of obstacles in the image for feature extraction, compares them with the prior model, extracts obstacles, and then uses the parallax information of the same object between different cameras to obtain The depth of the obstacle, so as to determine the position of the obstacle. The disadvantage of this method is that it consumes a lot of computing resources, is easily affected by illumination, and because the conversion error from image pixel...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

Application Information

Patent Timeline
no application Login to View More
Patent Type & Authority Patents(China)
IPC IPC(8): G05D1/02G01S17/931
CPCG05D1/0257G01S17/931
Inventor 张佳鹏项志宇
Owner ZHEJIANG UNIV