Accuracy of global navigation satellite system based positioning using high definition map based localization

• US 10,527,734 B2
• Filed: 11/21/2018
• Issued: 01/07/2020
• Est. Priority Date: 11/22/2017
• Status: Active Grant

• Current Assignee: NVIDIA Corporation
• Original Assignee: DeepMap, Inc. (NVIDIA Corporation)
• Inventors: Adachi, Jeffrey Minoru
• Primary Examiner(s): Liu, Harry K
• Application Number:
  US16/198,324
• Publication Number:
  US 20190154842A1
• Time in Patent Office: 412 Days
• Field of Search: 34235727
• CPC Class Codes: G01C 21/1652 with ranging devices, e.g. LIDAR or RADAR
• CPC Class Codes: G01C 21/1656 with passive imaging devices, e.g. cameras
• CPC Class Codes: G01C 21/30 Map- or contour-matching
• CPC Class Codes: G01C 21/36 Input/output arrangements for on-board computers
• CPC Class Codes: G01S 19/41 Differential correction, e.g. DGPS [differential GPS]
• CPC Class Codes: G01S 19/43 using carrier phase measurements, e.g. kinematic positioning; using long or short baseline interferometry
• CPC Class Codes: G01S 19/44 Carrier phase ambiguity resolution; Floating ambiguity; LAMBDA [Least-squares AMBiguity Decorrelation Adjustment] method
• CPC Class Codes: G01S 19/45 by combining measurements of signals from the satellite radio beacon positioning system with a supplementary measurement
• CPC Class Codes: G01S 19/485 whereby the further system is an optical system or imaging system
• CPC Class Codes: G01S 19/49 whereby the further system is an inertial position system, e.g. loosely-coupled

Abstract

A vehicle, for example, an autonomous vehicle receives signals from a global navigation satellite system (GNSS) and determines accurate location of the vehicle using the GNSS signal. The vehicle performs localization to determine the location of the vehicle as it drives. The autonomous vehicle uses sensor data and a high definition map to determine an accurate location of the autonomous vehicle. The autonomous vehicle uses accurate location of the vehicle to determine RTK corrections that is used for improving GNSS location estimates at a future location. The RTK corrections may be transmitted to other vehicles.

Claims

1. A non-transitory computer readable storage medium storing instructions, the instructions when executed by a processor, cause the processor to perform steps comprising:

initializing, by the processor of an autonomous vehicle configured to receive global navigation satellite system (GNSS) signals, a location of the autonomous vehicle and RTK corrections based on the location of the autonomous vehicle and raw GNSS signals from the location; and

repeating the steps comprising;

receiving, by a GNSS receiver mounted on the autonomous vehicle, a raw GNSS signal from the GNSS;

accessing previously determined RTK corrections by the autonomous vehicle;

determining an accurate GNSS position estimate by applying the previously determined RTK corrections to the raw GNSS signal;

receiving sensor data captured by sensors mounted on the autonomous vehicle;

accessing at least a portion of high definition (HD) map, the portion describing a geographical region surrounding the autonomous vehicle;

performing localization using the received sensor data and the accurate GNSS position estimate to determine a current location of the autonomous vehicle in the portion of the HD map;

determining new RTK corrections using the current location of the vehicle, wherein the new RT corrections are subsequently used for determining accurate GNSS position estimates;

determining control signals for navigating the autonomous vehicle using the current location of the autonomous vehicle and the data of the HD map; and

navigating the autonomous vehicle based on the control signals.

2. The non-transitory computer readable storage medium of claim 1, wherein the raw GNSS signal is a first GNSS signal, the sensor data is a first sensor data, the portion of the high definition map is a first portion of the high definition map, wherein instructions for initializing the location of the autonomous vehicle comprise instructions for:

receiving, by the autonomous vehicle, a second raw GNSS signal;

determining a GNSS position estimate based on the second raw GNSS signal;

receiving a second sensor data captured by sensors mounted on the autonomous vehicle;

accessing at least a second portion of high definition map, the second portion describing a geographical region surrounding the current location of the autonomous vehicle;

determining a first location of the vehicle based on at least the second portion of the high definition (HD) map, the second sensor data, and the GNSS position estimate; and

using the first location for initializing the location of the autonomous vehicle.

3. The non-transitory computer readable storage medium of claim 1, wherein instructions for determining the current location comprise instructions for:

building a first point cloud representing the geographical region surrounding the autonomous vehicle, the first point cloud built based on sensor data collected by sensors of the vehicle;

building a second point cloud based on the HD map; and

aligning the first point cloud to the second point cloud.

4. The non-transitory computer readable storage medium of claim 1, wherein instructions for determining the current location comprise instructions for:

determining one or more features describing the surrounding of the autonomous vehicle based on the sensor data; and

matching the one or more features with features in the HD map.

5. The non-transitory computer readable storage medium of claim 1, wherein instructions for determining the current location comprise instructions for:

determining an estimate of distance travelled from the previous location; and

determining the current location by combining the previous location and the estimate of the distance travelled from the previous location.

6. The non-transitory computer readable storage medium of claim 5, wherein the estimate of distance travelled from the previous location is determined using a measure of acceleration of the vehicle received from an inertial measurement unit (IMU) of the autonomous vehicle.

7. The non-transitory computer readable storage medium of claim 1, wherein the RTK corrections are determined by aggregating a plurality of previously determined RTK corrections.

8. The non-transitory computer readable storage medium of claim 1, wherein the sensor is a lidar mounted on the autonomous vehicle and the sensor data is a lidar scan captured by the lidar.

9. The non-transitory computer readable storage medium of claim 1, wherein the sensor is a camera mounted on the autonomous vehicle and the sensor data is a camera image captured by the camera.

10. The non-transitory computer readable storage medium of claim 1, wherein the instructions further cause the processor to perform the step of:

transmitting the RTK corrections to another autonomous vehicle for use by the other autonomous vehicle for determining the location of the other autonomous vehicle.

11. The non-transitory computer readable storage medium of claim 1, wherein the instructions further cause the processor to perform the step of:

transmitting the RTK corrections to another vehicle for determining a relative location between the autonomous vehicle and the other autonomous vehicle.

12. The non-transitory computer readable storage medium of claim 1, wherein the instructions further cause the processor to perform the step of:

receiving a point cloud from another autonomous vehicle;

aggregating a point cloud based on sensor data of the autonomous vehicle and the point cloud received from the other autonomous vehicle, the aggregating using a relative location between the autonomous vehicle and the other autonomous vehicle as a constraint; and

aligning the aggregate point cloud with a point cloud based on the data of the HD map to determine location of the vehicle.

13. A computer-implemented method comprising:

initializing, by a processor of a vehicle configured to receive global navigation satellite system (GNSS) signals, a location of the vehicle and RTK corrections based on the location of the vehicle and raw GNSS signals from the location; and

repeating steps comprising;

receiving, by a GNSS receiver mounted on the vehicle, a raw GNSS signal from the GNSS;

accessing previously determined RTK corrections by the vehicle;

determining an accurate GNSS position estimate by applying the previously determined RTK corrections to the raw GNSS signal;

receiving sensor data captured by sensors mounted on the vehicle;

accessing at least a portion of high definition (HD) map, the portion describing a geographical region surrounding the vehicle;

performing localization using the received sensor data and the accurate GNSS position estimate to determine a current location of the vehicle in the portion of the HD map; and

determining new RTK corrections using the current location of the vehicle, wherein the new RT corrections are subsequently used for determining accurate GNSS position estimates.

14. The computer-implemented method of claim 13, wherein the raw GNSS signal is a first GNSS signal, the sensor data is a first sensor data, the portion of the high definition map is a first portion of the high definition map, wherein initializing the location of the vehicle comprises:

receiving, by the vehicle, a second raw GNSS signal;

determining a GNSS position estimate based on the second raw GNSS signal;

receiving a second sensor data captured by sensors mounted on the vehicle;

accessing at least a second portion of high definition map, the second portion describing a geographical region surrounding the current location of the vehicle;

determining a first location of the vehicle based on at least the second portion of the high definition (HD) map, the second sensor data, and the GNSS position estimate; and

using the first location for initializing the location of the vehicle.

15. The computer-implemented method of claim 13, wherein determining the current location comprises:

building a first point cloud representing the geographical region surrounding the vehicle, the first point cloud built based on sensor data collected by sensors of the vehicle;

building a second point cloud based on the HD map; and

aligning the first point cloud to the second point cloud.

16. The computer-implemented method of claim 13, wherein determining the current location comprises:

determining one or more features describing the surrounding of the vehicle based on the sensor data; and

matching the one or more features with features in data of the HD map.

17. The computer-implemented method of claim 13, wherein determining the current location comprises:

determining an estimate of distance travelled from the previous location; and

determining the current location by combining the previous location and the estimate of the distance travelled from the previous location.

18. The computer-implemented method of claim 17, wherein the estimate of distance travelled from the previous location is determined using a measure of acceleration of the vehicle received from an inertial measurement unit (IMU) of the vehicle.

19. The computer-implemented method of claim 13, wherein the RTK corrections are determined by aggregating RTK corrections at a plurality of previous locations.

20. The computer-implemented method of claim 13, wherein the sensor is a lidar mounted on the vehicle and the sensor data is a lidar scan captured by the lidar.

21. The computer-implemented method of claim 13, wherein the sensor is a camera mounted on the vehicle and the sensor data is a camera image captured by the camera.

22. The computer-implemented method of claim 13, further comprising:

broadcasting the RTK corrections for use by other vehicles.

23. The computer-implemented method of claim 13, further comprising:

transmitting the RTK corrections to another vehicle for determining a relative location between the vehicle and the other vehicle.

24. The computer-implemented method of claim 13, further comprising:

receiving a point cloud from another vehicle;

aggregating a point cloud based on sensor data of the vehicle and the point cloud received from the other vehicle, the aggregating using a relative location between the vehicle and the other vehicle as a constraint; and

aligning the aggregate point cloud with a point cloud based on the data of the HD map to determine location of the vehicle.

25. A computer system comprising:

a processor; and

non-transitory computer readable storage medium storing instructions, the instructions when executed by the processor, cause the processor to perform steps comprising;

initializing, by the processor of an autonomous vehicle configured to receive global navigation satellite system (GNSS) signals, a location of the autonomous vehicle and RTK corrections based on the location of the autonomous vehicle and raw GNSS signals from the location; and

repeating the steps comprising;

receiving, by a GNSS receiver mounted on the autonomous vehicle, a raw GNSS signal from the GNSS;

accessing previously determined RTK corrections by the autonomous vehicle;

determining an accurate GNSS position estimate by applying the previously determined RTK corrections to the raw GNSS signal;

receiving sensor data captured by sensors mounted on the autonomous vehicle;

accessing at least a portion of high definition (HD) map, the portion describing a geographical region surrounding the autonomous vehicle;

performing localization using the received sensor data and the accurate GNSS position estimate to determine a current location of the autonomous vehicle in the portion of the HD map;

determining new RTK corrections using the current location of the vehicle, wherein the new RT corrections are subsequently used for determining accurate GNSS position estimates;

determining control signals for navigating the autonomous vehicle using the current location of the autonomous vehicle and the data of the HD map; and

navigating the autonomous vehicle based on the control signals.

26. The computer system of claim 25, wherein the raw GNSS signal is a first GNSS signal, the sensor data is a first sensor data, the portion of the high definition map is a first portion of the high definition map, wherein instructions for initializing the location of the autonomous vehicle comprise instructions for:

receiving, by the autonomous vehicle, a second raw GNSS signal;

determining a GNSS position estimate based on the second raw GNSS signal;

receiving a second sensor data captured by sensors mounted on the autonomous vehicle;

accessing at least a second portion of high definition map, the second portion describing a geographical region surrounding the current location of the autonomous vehicle;

determining a first location of the vehicle based on at least the second portion of the high definition (HD) map, the second sensor data, and the GNSS position estimate; and

using the first location for initializing the location of the autonomous vehicle.

27. The computer system of claim 25, wherein instructions for determining the current location comprise instructions for:

building a first point cloud representing the geographical region surrounding the autonomous vehicle, the first point cloud built based on sensor data collected by sensors of the vehicle;

building a second point cloud based on the HD map; and

aligning the first point cloud to the second point cloud.

28. The computer system of claim 25, wherein instructions for determining the current location comprise instructions for:

determining one or more features describing the surrounding of the autonomous vehicle based on the sensor data; and

matching the one or more features with features in data of the HD map.

29. The computer system of claim 25, wherein instructions for determining the current location comprise instructions for:

determining an estimate of distance travelled from the previous location; and

determining the current location by combining the previous location and the estimate of the distance travelled from the previous location.

30. The computer system of claim 29, wherein the estimate of distance travelled from the previous location is determined using a measure of acceleration of the vehicle received from an inertial measurement unit (IMU) of the autonomous vehicle.

31. The computer system of claim 25, wherein the RTK corrections are determined by aggregating RTK corrections at a plurality of previous locations.

32. The computer system of claim 25, wherein the sensor is a lidar mounted on the autonomous vehicle and the sensor data is a lidar scan captured by the lidar.

33. The computer system of claim 25, wherein the sensor is a camera mounted on the autonomous vehicle and the sensor data is a camera image captured by the camera.

34. The computer system of claim 25, wherein the instructions further cause the processor to perform the step of:

broadcasting the RTK corrections for use by other vehicles.

35. The computer system of claim 25, wherein the instructions further cause the processor to perform the step of:

transmitting the RTK corrections to another vehicle for determining a relative location between the autonomous vehicle and the other autonomous vehicle.

36. The computer system of claim 25, wherein the instructions further cause the processor to perform the step of:

receiving a point cloud from another autonomous vehicle;

aggregating a point cloud based on sensor data of the autonomous vehicle and the point cloud received from the other autonomous vehicle, the aggregating using a relative location between the autonomous vehicle and the other autonomous vehicle as a constraint; and

aligning the aggregate point cloud with a point cloud based on the data of the HD map to determine location of the vehicle.