Methods and Apparatus for Selecting a Map for a Moving Object, System, and Vehicle/Robot

Patent metadata

Field	Value
Publication number	US-2020116517-A1
Application number	US-201916714213-A
Country	US
Kind code	A1
Filing date	Dec 13, 2019
Priority date	Jun 16, 2017
Publication date	Apr 16, 2020
Grant date	—

Abstract

A method and apparatus for selecting a map from a plurality of maps having different resolutions for a moving object includes dynamically selecting an appropriate map from maps having different levels of details according to environment information, such that a minimum amount of required map is loaded, thereby effectively improving the data processing efficiency of vehicles.

First claim

What is claimed is:

1 . A computer-implemented method for selecting a map from a plurality of maps having different resolutions for a moving object, the method comprising:
obtaining localization information of the moving object; receiving sensor data from a sensor; calculating an uncertainty value representing an uncertainty of the sensor data; and selecting the map from the plurality of maps having the different resolutions according to the localization information and the uncertainty value representing the uncertainty of the sensor data.

2 . The method according to claim 1 , wherein the localization information of the moving object is represented by a pose of the moving object, wherein the pose includes a two dimensional location and a heading angle.

3 . The method according to claim 2 , wherein the uncertainty value representing the uncertainty of the sensor data being calculated according to the pose of the moving object.

4 . The method according to claim 3 , wherein calculating the uncertainty value further comprises:
calculating a pose of the sensor data in a map coordinate system; constructing a covariance matrix of the pose of the sensor data; calculating a maximum eigenvalue of the covariance matrix; and setting the uncertainty value representing the uncertainty of the sensor data to be the maximum eigenvalue.

5 . The method according to claim 4 , wherein the pose of the sensor data

P
=

[

x

y

θ

]

is denoted as:
P=RP̆+P S
wherein, x, y are P's x-y coordinates in the map coordinate system and θ is the heading angle, wherein the raw observation of P in the moving object coordinate system is denoted as

P
⋓

=

[

x
⋓

y
⋓

θ
⋓

]

,

and wherein the pose of the moving object in the map coordinate system is denoted as

P
S

=

[

x
s

y
s

θ
s

]

,

where

R
=

[

cos




θ
s

-
sin





θ
s

0

sin




θ
s

cos




θ
s

0

0

0

1

]

;

denoting P's covariance matrix Σ as:
Σ= J 1 Σ̆J 1 T +J 2 Σ S J 2 T

wherein, Σ̆ is P̆'s covariance matrix, Σ S is P S 's covariance matrix, J 1 =R, and

J
2

=

[

1

0

-
sin





θ
s



Assignees

• Bayerische Motoren Werke Ag

Inventors

• Doemling Maximilian
• Granzow Sebastian
• Jiang Wanli
• Li Qianshan
• Li Jianpeng
• Lv Shuhan
• Xu Hongshan
• Xu Tao

Classifications

• G01C21/20

  Instruments for performing navigational calculations (G01C21/24, G01C21/26 take precedence) · CPC title

• G01C21/28 Primary

  with correlation of data from several navigational instruments · CPC title

• G05D1/0253

  extracting relative motion information from a plurality of images taken successively, e.g. visual odometry, optical flow (determining position or orientation from images G06T7/70) · CPC title

• G05D1/0255

  using acoustic signals, e.g. ultra-sonic singals (sonar systems designed for anti-collision purposes G01S15/93) · CPC title

• G06F17/16

  Matrix or vector computation {, e.g. matrix-matrix or matrix-vector multiplication, matrix factorization (matrix transposition G06F7/78)} · CPC title