Unmanned vehicle, method, apparatus and system for positioning unmanned vehicle

Patent metadata

Field	Value
Publication number	US-10162358-B2
Application number	US-201615282984-A
Country	US
Kind code	B2
Filing date	Sep 30, 2016
Priority date	May 24, 2016
Publication date	Dec 25, 2018
Grant date	Dec 25, 2018

Abstract

Disclosed embodiments include an unmanned vehicle, a method, apparatus and system for positioning an unmanned vehicle. In some embodiments, the method includes: acquiring first laser point cloud height value data matching a current position of the unmanned vehicle, the first laser point cloud height value data; converting the first laser point cloud height value data into laser point cloud projection data in a horizontal earth plane; determining a first matching probability of the laser point cloud projection data in a predetermined range of a laser point cloud height value map; and determining a position of the unmanned vehicle in the laser point cloud height value map based on the first matching probability. The embodiment implements an accurate positioning on the current position of the unmanned vehicle.

First claim

What is claimed is:

1. A method for positioning an unmanned vehicle based on a laser point cloud height value matching, comprising:
irradiating objects around a vehicle with a laser, and generating reflected laser points from the irradiated objects to create a laser point cloud, each reflected laser point having coordinates and a height value;
acquiring, by a computer processor, first laser point cloud height value data matching a current position of the unmanned vehicle;
converting the first laser point cloud height value data into laser point cloud projection data in a horizontal earth plane;
determining a first matching probability of the laser point cloud projection data in a predetermined range of a laser point cloud height value map by using a position of a predetermined prior positioning position in the laser point cloud height value map as an initial position, the laser point cloud height value map being generated by:
dividing the Earth surface into M×N map areas in a horizontal earth plane of a world coordinate system;
dividing each of the map areas into m×n map grids;
collecting second laser point cloud height value data corresponding to positioning positions of each of the map grids, the second laser point cloud height value data comprising second coordinates of laser points in the world coordinate system and height values corresponding to the laser points within the second laser point cloud height value data; and
storing corresponding map data in each of the map grids, the map data comprising an average of height values of laser points in a positioning position corresponding to the map grid and a number of the laser points in the positioning position corresponding to the map grid; and

determining a position of the unmanned vehicle in the laser point cloud height value map based on the first matching probability.

2. The method according to claim 1 , wherein the map areas have an identical size and shape.

3. The method according to claim 1 , wherein a first coordinate of each of the laser points is a coordinate of each of the laser points within the first laser point cloud height value data, in a vehicle coordinate system of the unmanned vehicle;
the converting the first laser point cloud height value data into laser point cloud projection data in a horizontal earth plane comprising:
converting the first laser point cloud height value data into third laser point cloud height value data, the third laser point cloud height value data comprising third coordinates of laser points and height values corresponding to the laser points within the first laser point cloud height value data, the third coordinate being a coordinate of each of the laser points within the first laser point cloud height value data in the world coordinate system; and
projecting the third laser point cloud height value data to the horizontal earth plane to generate the laser point cloud projection data.

4. The method according to claim 3 , wherein the third coordinate X′ is defined as:
X ′=( x′,y′,z ′) T =RX+T;

where R is a rotation matrix for a transformation from the vehicle coordinate system of the unmanned vehicle to the world coordinate system, X=(x, y, z) T is the first coordinate of each of the laser points within the first laser point cloud height value data, and T is a translation matrix for a transformation from the vehicle coordinate system of the unmanned vehicle to the world coordinate system.

5. The method according to claim 4 , wherein:
the laser point cloud projection data comprises projection coordinates of the laser points within the first laser point cloud height value data, an average of height values of laser points in projection grids, and a number of the laser points in the projection grids;
where the projection coordinate X″ of each of the laser points within the first laser point cloud height value data meets:
X ″=( x″,y ″) T =SX′;

 S being a projection matrix and meeting:

S
=

[

1

0

0

0

1

0

]

;

and
each of the projection grids having a size and shape identical to the size and shape of the map grid.

6. The method according to claim 5 , the determining a first matching probability of the laser point cloud projection data in a predetermined range of a laser point cloud height value map by using a position of a predetermined prior positioning position in the laser point cloud height value map as an initial position comprising:
overlapping a center projection grid O(x, y) of the laser point cloud projection data with a map grid corresponding to the prior positioning position O′(x o , y o ) in the laser point cloud height value map, wherein the center projection grid O(x, y) is a projection grid representing a vehicle body of the unmanned vehicle in the laser point cloud projection data;
determining a first matching probability of a projection range of the laser point cloud projection data and a corresponding map range;
moving the center projection grid O(x, y) by a predetermined offset k, and respectively determining the first matching probability of the laser point cloud projection data corresponding to the current center projection grid O(x, y); and
the determining a position of the unmanned vehicle in the laser point cloud height value map based on the first matching probability comprising:
determining a position of the unmanned vehicle in the laser point cloud height value map based on a weighted average of the first matching probability.

7. The method according to claim 6 , wherein:
the first matching probability P(x, y) corresponding to the center projection grid O(x, y) is defined as:

P
⁡

(

x
,
y

)

=

α
-

∑

x
i

=

x
1

x
m

⁢

∑

y
j

=

y

Assignees

• Baidu online network technology beijing co ltd

Inventors

• Song Shiyu
• Li Wenbo
• Zhang Tianlei

Classifications

• G06N7/01

  Probabilistic graphical models, e.g. probabilistic networks · CPC title

• G05D1/0274

  using mapping information stored in a memory device (navigation using map-matching G01C21/30) · CPC title

• G06N7/005

  Physics · mapped topic

• G01S7/4808

  Evaluating distance, position or velocity data · CPC title

• G05D2201/0213

  Physics · mapped topic