High definition map and route storage management system for autonomous vehicles

Patent metadata

Field	Value
Publication number	US-11775570-B2
Application number	US-202117381067-A
Country	US
Kind code	B2
Filing date	Jul 20, 2021
Priority date	Dec 30, 2016
Publication date	Oct 3, 2023
Grant date	Oct 3, 2023

Abstract

High definition maps for autonomous vehicles are very high resolution and detailed, and hence require storage of a great deal of data. A vehicle computing system provides multi-layered caching makes this data usable in a system that requires very low latency on every operation. The system determines which routes are most likely to be driven in the near future by the car, and ensures that the route is cached on the vehicle before beginning the route. The system provides efficient formats for moving map data from server to car and for managing the on-car disk. The system further provides real-time accessibility of nearby map data as the car moves, while providing data access at optimal speeds.

First claim

What is claimed is:

1. A method comprising:
storing a plurality of map tiles in a compressed format;
loading, based at least on a determined route for a machine, a first subset of map tiles from the plurality of map tiles into a storage cache in a decompressed format;
loading, based at least on a predicted movement corresponding to the machine, a second subset of map tiles from the first subset of map tiles from the storage cache into random access memory (RAM); and
performing one or more navigation operations by the machine based at least on the second subset of map tiles stored in the RAM.

2. The method of claim 1 , wherein the predicted movement is within a threshold time interval from a point in time at which the machine is at a determined location.

3. The method of claim 2 , wherein the determined location is based at least on one or more of:
sensor data obtained using one or more sensors corresponding to the machine; or one or more map tiles previously loaded to the RAM.

4. The method of claim 1 , wherein the predicted movement is based at least on a plurality of previous locations corresponding to the machine over a particular amount of time.

5. The method of claim 1 , wherein at least one map tile of the second subset of map tiles corresponds to an area that is adjacent to a current area corresponding to a current map tile that includes a determined location of the machine.

6. A processor comprising:
processing circuitry to cause performance of operations comprising:
storing a plurality of map tiles in a compressed format in first storage location;
loading a first subset of map tiles from the plurality of map tiles from the first storage location to a second storage location in a decompressed format;
loading a second subset of map tiles of the first subset of map tiles from the second storage location to a third storage location based at least on the second subset of map tiles corresponding to a predicted movement corresponding to a machine, the third storage location corresponding to a temporary memory storage; and
performing one or more operations by the machine based at least on the second subset of map tiles stored in the third storage location.

7. The processor of claim 6 , wherein the first subset of map tiles are selected for storage in the second storage location based at least on the first subset of map tiles corresponding to a determined route corresponding to the machine.

8. The processor of claim 6 , wherein the predicted movement is determined based at least on a determined location corresponding to the machine.

9. The processor of claim 8 , wherein the determined location is determined based at least on one or more of:
sensor data obtained using one or more sensors corresponding to the machine; or one or more map tiles previously loaded to the third storage location.

10. The processor of claim 6 , wherein the predicted movement is within a threshold time interval from a point in time at which the machine is at a location.

11. The processor of claim 6 , wherein the predicted movement is based at least on localization data corresponding to the machine and corresponding to a plurality of points in time.

12. The processor of claim 6 , wherein the third storage location has faster accessibility than the second storage location, and the second storage location has faster accessibility than the first storage location.

13. The processor of claim 6 , wherein the third storage location comprises random access memory (RAM).

14. A system comprising:
one or more processing units to cause performance of operations comprising:
loading, based at least on a determined route of a machine, a first subset of map tiles stored in a compressed format into a first temporary storage location in a decompressed format;
loading, based at least on predicted movement of the machine from a current location to a determined location along a portion of the determined route, a second subset of map tiles from the first subset of map tiles to a second temporary storage location; and
accessing the second subset of map tiles during navigation of the machine along at least the portion of the determined route.

15. The system of claim 14 , wherein the first subset of map tiles and the second subset of map tiles are included a set of map tiles corresponding to a high definition (HD) map.

16. The system of claim 14 , wherein the second temporary storage location is associated with faster access than the first temporary storage location.

17. The system of claim 14 , wherein at least one tile in the first subset of map tiles is associated with one or more locations further from a current location of the machine than the second subset of map tiles.

18. The system of claim 14 , wherein the predicted movement is within a threshold time interval from a point in time at which the machine is at the current location.

19. The system of claim 14 , wherein the first temporary storage includes a cache and the second temporary storage includes random access memory (RAM).

20. The system of claim 14 , wherein the first subset of tiles and the second subset of tiles are stored in a compressed format prior to being loaded into the first temporary storage location in the decompressed format.

Assignees

• Deepmap Inc
• Nvidia Corp

Inventors

• Wheeler Mark Damon

Classifications

• G06F16/29 Primary

  Geographical information databases · CPC title

• G01C21/3881

  Tile-based structures · CPC title

• G01C21/3889

  Transmission of selected map data, e.g. depending on route · CPC title

• G05D1/0088

  characterized by the autonomous decision making process, e.g. artificial intelligence, predefined behaviours (using knowledge based models G06N5/00) · CPC title

• G05D1/0212

  with means for defining a desired trajectory (involving a plurality of land vehicles G05D1/0287) · CPC title