Radar multipath processing

• US 10,296,001 B2
• Filed: 10/27/2016
• Issued: 05/21/2019
• Est. Priority Date: 10/27/2016
• Status: Active Grant

• Current Assignee: Uatc, LLC (Uber Technologies, Inc.)
• Original Assignee: Uber Technologies, Inc.
• Inventors: Smith, Warren, Smith, Howard
• Primary Examiner(s): Heinle, Courtney D
• Assistant Examiner(s): Martinez Borrero, Luis A
• Application Number:
  US15/335,692
• Publication Number:
  US 20180120842A1
• Time in Patent Office: 936 Days
• CPC Class Codes: B60W 40/00 Estimation or calculation of non-directly measurable driving parameters for road vehicle drive control systems not related to the control of a particular sub unit, e.g. by using mathematical models
• CPC Class Codes: G01S 13/42 Simultaneous measurement of distance and other co-ordinates indirect measurement G01S13/46
• CPC Class Codes: G01S 13/865 Combination of radar systems with lidar systems
• CPC Class Codes: G01S 13/867 Combination of radar systems with cameras
• CPC Class Codes: G01S 13/931 of land vehicles
• CPC Class Codes: G01S 17/931 of land vehicles
• CPC Class Codes: G01S 7/412 based on a comparison between measured values and known or stored values

Abstract

A control system of an autonomous vehicle (AV) can receive sensor data from a sensor array of the AV. The sensor data can comprise raw radar data from a radar system of the sensor array, and in many examples, a live LIDAR data from a LIDAR system of the sensor array. In certain implementations the control system can access a current localization map or a live LIDAR map of a surrounding area of the AV, and compare the raw radar data with the current localization map or the live LIDAR map to identify multipath objects in the raw radar data. The control system may then remove the multipath objects or track the actual objects corresponding to the multipath objects accordingly.

Claims

1. An autonomous vehicle (AV) comprising:

a sensor array comprising at least a radar system;

acceleration, braking, and steering systems;

a memory storing a set of localization maps for a given region; and

an AV control system comprising one or more processors executing an instruction set causing the AV control system to;

receive sensor data from the sensor array, the sensor data including radar data from the radar system;

access a current localization map from the set of localization maps stored in the memory, the current localization map indicating a current surrounding of the AV;

compare the radar data with the current localization map to identify multipath objects in the radar data;

from the identified multipath objects, identify one or more multipath objects of interest;

track the one or more multipath objects of interest; and

analyze the current localization map, the sensor data, and the one or more multipath objects of interest to operate the acceleration, braking, and steering systems of the AV along a current route within the given region.

2. The AV of claim 1, wherein the executed instruction set further causes the AV control system to:

process the radar data by removing at least one of the identified multipath objects from the radar data using the current localization map, wherein the analyzed sensor data for operating the acceleration, braking, and steering systems includes the processed radar data.

3. The AV of claim 1, wherein the executed instruction set causes the AV control system to identify the one or more multipath objects of interest based on a velocity variance between the one or more multipath objects of interest and a remainder of the identified multipath objects.

4. The AV of claim 1, wherein the sensor array further comprises at least one of a LIDAR system and a camera system.

5. The AV of claim 4, wherein the executed instruction set further causes the AV control system to identify the one or more multipath objects of interest by comparing identified multipath objects with at least one of LIDAR data from the LIDAR system or image data from the camera system.

6. The AV of claim 4, wherein the executed instruction set causes the AV control system to track the one or more multipath objects of interest to at least partially resolve an occlusion in at least one of LIDAR data from the LIDAR system or image data from the camera system.

7. The AV of claim 6, wherein the executed instruction set causes the AV control system to track the one or more multipath objects of interest to at least partially resolve the occlusion by performing a number of ray tracing operations on the sensor data to determine a trajectory of each of the one or more multipath objects of interest.

8. A radar multipath processing system comprising:

one or more processors; and

one or more memory resources storing instructions that, when executed by the one or more processors, cause the one or more processors to;

receive radar data from a radar system of an autonomous vehicle (AV);

access a current localization map from a set of localization maps stored on-board the AV, the current localization map indicating a current surrounding of the AV; and

compare the radar data with the current localization map to identify multipath objects in the radar data;

from the identified multipath objects, identify one or more multipath objects of interest; and

track the one or more multipath objects of interest for the AV.

9. The radar multipath processing system of claim 8, wherein the executed instructions further cause the one or more processors to:

process the radar data by removing at least one of the identified multipath objects from the radar data using the current localization map; and

provide the processed radar data to an AV control system that utilizes the processed radar data to operate acceleration, braking, and steering systems of the AV along a given route.

10. The radar multipath processing system of claim 8, wherein the executed instructions cause the one or more processors to identify the one or more multipath objects of interest based on a velocity variance between the one or more multipath objects of interest and a remainder of the identified multipath objects.

11. The radar multipath processing system of claim 8, wherein the executed instructions cause the one or more processors to identify the one or more multipath objects of interest by comparing the identified multipath objects with at least one of LIDAR data from a LIDAR system of the AV or image data from a camera system of the AV.

12. The radar multipath processing system of claim 8, wherein the executed instructions cause the one or more processors to track the one or more multipath objects of interest to at least partially resolve an occlusion in at least one of LIDAR data from a LIDAR system of the AV or image data from a camera system of the AV.

13. The radar multipath processing system of claim 12, wherein the executed instructions cause the one or more processors to track the one or more multipath objects of interest to at least partially resolve the occlusion by performing a number of ray tracing operations on the radar data to determine a trajectory of each of the one or more multipath objects of interest.

14. A computer-implemented method of autonomously operating a vehicle, the method being performed by one or more processors and comprising:

receiving sensor data from a sensor array of the vehicle, the sensor data including radar data from a radar system of the vehicle;

accessing a current localization map for a given region from a set of localization maps stored in memory on-board the vehicle, the current localization map indicating a current surrounding of the vehicle;

comparing the radar data with the current localization map to identify multipath objects in the radar data;

from the identified multipath objects, identify one or more multipath objects of interest; and

track the one or more multipath objects of interest for the vehicle,

analyzing the current localization map, the sensor data, and the one or more multipath objects of interest to operate acceleration, braking, and steering systems of the vehicle along a current route within the given region.

15. The method of claim 14, further comprising:

processing the radar data by removing at least one of the identified multipath objects from the radar data using the current localization map, wherein the analyzed sensor data for operating the acceleration, braking, and steering systems includes the processed radar data.

16. The method of claim 14, wherein the one or more processors identify the one or more multipath objects of interest based on a velocity variance between the one or more multipath objects of interest and a remainder of the identified multipath objects.

17. The method of claim 14, wherein the sensor array further comprises at least one of a LIDAR system and a camera system.