Map generation systems and methods

Patent metadata

Field	Value
Publication number	US-11480435-B2
Application number	US-201916703550-A
Country	US
Kind code	B2
Filing date	Dec 4, 2019
Priority date	Jun 30, 2017
Publication date	Oct 25, 2022
Grant date	Oct 25, 2022

Abstract

A method of map generation includes receiving data from a plurality of vehicles about environments within which the plurality of vehicles operate, and generating a three-dimensional map using the data from the plurality of vehicles. The data is collected by one or more sensors on-board the plurality of vehicles.

First claim

What is claimed is:

1. A method of map generation comprising:
receiving data from a plurality of vehicles about environments within which the plurality of vehicles operate, wherein the data is collected by one or more sensors on-board the plurality of vehicles; and
generating, with aid of one or more processors, a three-dimensional map using the data from the plurality of vehicles, including:
determining and recognizing, from the data from the plurality of vehicles, dissimilar data provided by one or more vehicles of the plurality of vehicles;
determining that one vehicle of the one or more vehicles has a vehicle positioning error, obtaining a correct position of the one vehicle by calibrating the positioning error of the one vehicle; and
generating, using the dissimilar data, a portion of the three-dimensional map associated with the correct position of the one vehicle.

2. The method of claim 1 , further comprising providing information useful for navigation to at least one vehicle of the plurality of vehicles, wherein the information is derived from the three-dimensional map.

3. The method of claim 2 , wherein the information useful for navigation comprises a subset of the three-dimensional map comprising an environment within which the at least one vehicle is operating.

4. The method of claim 1 , further comprising modifying a route of at least one vehicle of the plurality of vehicles to a specified destination based on the collected data or provided on the three-dimensional map.

5. The method of claim 1 , wherein the three-dimensional map is generated with aid of a cloud server or is stored in memory using cloud computing infrastructure.

6. The method of claim 1 , wherein the three-dimensional map comprises point cloud data.

7. The method of claim 6 , further comprising projecting the point cloud data to a world coordinate system.

8. The method of claim 1 , wherein the three-dimensional map further comprises image data collected by the plurality of vehicles.

9. The method of claim 1 , further comprising associating a confidence measure to the data that is used for generating a portion of the three-dimensional map.

10. The method of claim 9 , further comprising determining a degree of similarity between the data provided by one or more vehicles of the plurality of vehicles for generating the portion of the three-dimensional map.

11. The method of claim 9 , further comprising providing a higher confidence measure to similar data and providing a lower confidence measure to the dissimilar data.

12. The method of claim 11 , further comprising selecting data with a highest confidence measure for generating the portion of the three-dimensional map.

13. The method of claim 1 , wherein the vehicle positioning error is detected based on reliability of Global Position System (GPS) information.

14. The method of claim 13 , further comprising discarding the dissimilar data that is provided by the one vehicle having the vehicle positioning error.

15. The method of claim 1 , wherein the data collected by the one or more sensors comprises label information associated with a plurality of objects in the map.

16. The method of claim 15 , wherein the label information comprises class information and/or motion information associated with the plurality of objects in the three-dimensional map.

17. The method of claim 16 , wherein the plurality of objects comprise dynamic objects and/or static objects.

18. The method of claim 1 , further comprising:
determining confidence measure for the data from the plurality of vehicles based on:
at least one of a condition of each individual vehicle of the plurality of vehicles or an environment within which each individual vehicle operates; and
comparisons among the data from the plurality of vehicles.

19. A system of generating a map comprising:
one or more communication units; and
one or more processors configured to:
obtain, through the one or more communication units, data from a plurality of vehicles about environments within which the plurality of vehicles operate, wherein the data is collected by one or more sensors on-board the plurality of vehicles; and
generate a three-dimensional map using the data from the plurality of vehicles, including:
determining and recognizing, from the data from the plurality of vehicles, dissimilar data provided by one or more vehicles of the plurality of vehicles;
determining that one vehicle of the one or more vehicles has a vehicle positioning error, obtaining a correct position of the one vehicle by calibrating the positioning error of the one vehicle; and
generating, using dissimilar data, a portion of the three-dimensional map associated with a correct position of the one vehicle.

20. A method of map generation comprising:
receiving data from a plurality of vehicles about environments within which the plurality of vehicles operate, wherein the data is collected by one or more sensors on-board the plurality of vehicles; and
generating, with aid of one or more processors, a three-dimensional map using the data from the plurality of vehicles, including:
associating a confidence measure to for generating a portion of the three-dimensional map;
determining a degree of similarity between the data provided by one or more vehicles of the plurality of vehicles for generating the portion of the three-dimensional map;
determining that one vehicle of the one or more vehicles that provides dissimilar data has a vehicle positioning error, obtaining a correct position of the one vehicle by calibrating the positioning error of the one vehicle; and
using the dissimilar data for generating a portion of the three-dimensional map associated with the correct position of the one vehicle.

Assignees

• Sz Dji Technology Co Ltd

Inventors

• Li Wei
• Ma Lu

Classifications

• G08G1/0112

  from the vehicle, e.g. floating car data [FCD] · CPC title

• G06V20/56

  exterior to a vehicle by using sensors mounted on the vehicle · CPC title

• G01C21/3804

  Creation or updating of map data · CPC title

• G05D1/0278

  using satellite positioning signals, e.g. GPS · CPC title

• G05D2201/0213

  Physics · mapped topic