Obstacle detecting method and obstacle detecting apparatus based on unmanned vehicle, and device, and storage medium

Patent metadata

Field	Value
Publication number	US-11043002-B2
Application number	US-201916512940-A
Country	US
Kind code	B2
Filing date	Jul 16, 2019
Priority date	Sep 7, 2018
Publication date	Jun 22, 2021
Grant date	Jun 22, 2021

Abstract

The application provides an obstacle detecting method and obstacle detecting apparatus based on an unmanned vehicle, and a device, and a storage medium, where the method includes obtaining point cloud data collected by a detecting device, projecting the point cloud data onto a two-dimensional plane to obtain a two-dimensional projection grid graph, where the two-dimensional projection grid graph has multiple grids and two-dimensional data, and two-dimensional data is data obtained after the point cloud data is projected; generating multiple straight lines according to the two-dimensional projection grid graph, where each of the multiple straight lines has two-dimensional data, and each straight line has parameter information which represents the relationship between the straight line and other straight lines in the multiple straight lines; and determining orientation information of the obstacle according to the two-dimensional data and the parameter information of each of the multiple straight lines.

First claim

What is claimed is:

1. An obstacle detecting method based on an unmanned vehicle, comprising:
obtaining point cloud data collected by a detecting device, wherein the point cloud data is data obtained after an obstacle around the unmanned vehicle is detected by the detecting device;
projecting the point cloud data onto a two-dimensional plane to obtain a two-dimensional projection grid graph, wherein the two-dimensional projection grid graph has multiple grids and two-dimensional data, and the two-dimensional data is data obtained after the point cloud data is projected;
generating multiple straight lines according to the two-dimensional projection grid graph, wherein each of the multiple straight lines has at least one of two-dimensional data, and the each straight line has parameter information which represents a relationship between the straight line and other straight lines in the multiple straight lines; and
determining orientation information of the obstacle according to the two-dimensional data and the parameter information of the each of the multiple straight lines;
wherein the generating multiple straight lines according to the two-dimensional projection grid graph comprises:
for each grid of the two-dimensional projection grid graph, if there is two-dimensional data in the grid, assigning the grid a first value;
for each grid of the two-dimensional projection grid graph, if there is no two-dimensional data in the grid, assigning the grid a second value; and
generating the multiple straight lines according to the grid with the first value.

2. The method according to claim 1 , wherein the projecting the point cloud data onto a two-dimensional plane to obtain a two-dimensional projection grid graph comprises:
projecting the point cloud data onto the two-dimensional plane to obtain a two-dimensional projection graph, wherein the two-dimensional projection graph has the two-dimensional data;
performing gridding processing on the two-dimensional projection graph to obtain the two-dimensional projection grid graph.

3. The method according to claim 1 , wherein the for each grid of the two-dimensional projection grid graph, if there is two-dimensional data in the grid, assigning the grid a first value comprises:
for each grid of the two-dimensional projected grid graph, if there is the two-dimensional data in the grid, assigning the grid the first value according to a number of two-dimensional data in the grid, wherein the number of two-dimensional data is proportional to the first value.

4. The method according to claim 1 , wherein the generating the multiple straight lines according to the grid with the first value comprises:
connecting at least two grids with the first value as a straight line to generate the multiple straight lines.

5. The method according to claim 1 , wherein the generating the multiple straight lines according to the grid with the first value comprises:
performing Hough line detection on the two-dimensional projection grid graph according to the grid with the first value, to obtain the multiple straight lines.

6. The method according to claim 1 , wherein the determining orientation information of the obstacle according to the two-dimensional data and the parameter information of the each of the multiple straight lines comprises:
determining, according to the two-dimensional data and parameter information of the each of the multiple straight lines, an orientation confidence value of the each straight line; and
determining direction information indicated by a straight line having a largest orientation confidence value as the orientation information of the obstacle.

7. The method according to claim 6 , wherein the determining, according to the two-dimensional data and parameter information of the each of the multiple straight lines, an orientation confidence value of the each straight line comprises:
determining, for the each straight line, an intersecting straight line intersecting with the each straight line, wherein the parameter information of the each straight line is an angle and an intersection point between the each straight line and each intersecting straight line corresponding to the each straight line; and
determining an angle parameter of the each straight line according to angle values of all angles of the each straight line;
determining a position parameter of the each straight line according to position information of all intersecting points of the each straight line;
determining a distance between a first grid on the each straight line and the last grid on the each straight line as a distance parameter of the each straight line; and
determining the orientation confidence value of the each straight line according to the number of two-dimensional data on the each straight line, the angle parameter, the position parameter, and the distance parameter.

8. The method according to claim 7 , wherein the determining an angle parameter of the each straight line according to angle values of all angles of the each straight line comprises:
obtaining the angle parameter of the each straight line by performing a weighted summation on the angle values of all the angles of the each straight line.

9. The method according to claim 7 , wherein the determining a position parameter of the each straight line according to position information of all intersecting points of the each straight line comprises:
obtaining the position parameter of the each straight line by performing a weighted summation on the position information of all the intersecting points of the each straight line.

10. The method according to claim 7 , wherein the determining the orientation confidence value of the each straight line according to the number of two-dimensional data on the each straight line, the angle parameter, the position parameter, and the distance parameter comprises:
obtaining the orientation confidence value of the each straight line by performing a weighted summation on the number of two-dimensional data on the each straight line, the angle parameter, the position parameter, and the distance parameter.

11. An obstacle detecting apparatus based on an unmanned vehicle, comprising: a processor and a computer readable medium for storing program codes, which, when executed by the processor, cause the processor to:
obtain point cloud data collected by a detecting device, wherein the point cloud data is data obtained after an obstacle around the unmanned vehicle is detected by the detecting device;
project the point cloud data onto a two-dimensional plane to obtain a two-dimensional projection grid graph, wherein the two-dimensional projection grid graph has multiple grids and two-dimensional data, and the two-dimensional data is data obtained after the point cloud data is projected;
generate multiple straight lines according to the two-dimensional projection grid graph, wherein each of the multiple straight lines has at least one of two-dimensional data, and the each straight line has parameter information which represents a relationship between the straight line and other straight lines in the multiple straight lines; and
determine orientation information of the obstacle according to the two-dimensional data and the parameter information of the each of the multiple straight lines;
wherein the program codes further cause the processor to:
for each grid of the two-dimensional projection grid graph, if there is two-dimensional data in the grid, assign the grid a first value;
for each grid of the two-dimensional projection grid graph, if there is no two-dimensional data in the grid, assign the grid a second value; and
generate the multiple straight lines according to

Assignees

• Baidu online network technology beijing co ltd
• Baidu Online Network Tech Co Ltd

Inventors

• Guo Jiang

Classifications

• G06V10/48

  by mapping characteristic values of the pattern into a parameter space, e.g. Hough transformation · CPC title

• G01S7/4802 Primary

  using analysis of echo signal for target characterisation; Target signature; Target cross-section · CPC title

• G06T7/70 Primary

  Determining position or orientation of objects or cameras (camera calibration G06T7/80) · CPC title

• G06V20/58 Primary

  Recognition of moving objects or obstacles, e.g. vehicles or pedestrians; Recognition of traffic objects, e.g. traffic signs, traffic lights or roads · CPC title

• G06T2207/20061

  Hough transform · CPC title