Eliminating sensor self-hit data

Patent metadata

Field	Value
Publication number	US-11726188-B2
Application number	US-202016855478-A
Country	US
Kind code	B2
Filing date	Apr 22, 2020
Priority date	Apr 22, 2020
Publication date	Aug 15, 2023
Grant date	Aug 15, 2023

Abstract

The subject disclosure relates to ways to identify self-hit data collected by autonomous vehicle (AV) sensors. In some aspects, a method of the disclosed technology includes steps for generating a geometric model of an autonomous vehicle (AV), wherein the geometric model specifies physical boundaries of the AV in three-dimensional (3D) space, collecting sensor data for an environment around the AV, and identifying one or more data points, from among the collected sensor data, that correspond with a surface of the AV. Systems and machine-readable media are also provided.

First claim

What is claimed is:

1. A computer-implemented method, comprising:
collecting first sensor data for an environment around an autonomous vehicle (AV) at a first time;
as the AV navigates, dynamically creating a geometric model of the AV, wherein the geometric model specifies physical boundaries of the AV in three-dimensional (3D) space captured by the first sensor data;
creating a first mask from the geometric model that distinguishes a first set of sensor data that corresponds with a surface of the AV from a second set of sensor data that corresponds with environmental data;
filtering out, based on an application of the first mask to the first sensor data, data points from among the first set of sensor data that corresponds with the surface of the AV;
collecting second sensor data for the environment around the AV at a second time;
updating the geometric model to create a second mask based on the second sensor data, wherein the second mask is modified in accordance with the second sensor data;
filtering out, based on an application of the second mask to the second sensor data, a third set of data points from among the second sensor data that corresponds with the surface of the AV at the second time; and
navigating the AV based on the filtered second sensor data to exclude the surface of the AV.

2. The computer-implemented method of claim 1 , further comprising:
dynamically identifying the third set of data points from among the second sensor data as self-hit data points instead of a surrounding environment of the AV: and
categorizing the self-hit data points within the second sensor data.

3. The computer-implemented method of claim 1 , further comprising:
collecting Light Detection and Ranging (LiDAR) test data for the environment around the AV;
generating the first mask, wherein the first mask is a filter mask based on the geometric model that is created based on the LiDAR test data, and wherein identifying the first set of sensor data that corresponds with the surface of the AV is based on the filter mask.

4. The computer-implemented method of claim 1 , wherein the geometric model is based on a combination of a Computer-aided Design (CAD) model of the AV and the second sensor data.

5. The computer-implemented method of claim 1 , wherein the second sensor data comprises Light Detection and Ranging (LiDAR) data.

6. The computer-implemented method of claim 1 , further comprising:
collecting test data for the environment around the AV, and wherein the geometric model of the AV is based on the test data.

7. The computer-implemented method of claim 1 , wherein the geometric model comprises one or more buffers to facilitate use of the geometric model on different vehicle types.

8. A system for detecting sensor data self-hits, the system comprising:
one or more processors;
at least one Light Detection and Ranging (LiDAR) sensor coupled to the one or more processors; and
a computer-readable medium comprising instructions stored therein, which when executed by the processors, cause the processors to perform operations comprising:
collecting first sensor data for an environment around an autonomous vehicle (AV) at a first time;
as the AV navigates, dynamically creating a geometric model of the AV, wherein the geometric model specifies physical boundaries of the AV in three-dimensional (3D) space captured by the first sensor data;
creating a first mask from the geometric model that distinguishes a first set of sensor data that corresponds with a surface of the AV from a second set of sensor data that corresponds with environmental data;
filtering out, based on an application of the first mask to the first sensor data, data points from among the first set of sensor data that corresponds with the surface of the AV;
collecting second sensor data for the environment around the AV at a second time;
updating the geometric model to create a second mask based on the second sensor data, wherein the second mask is modified in accordance with the second sensor data;
filtering out, based on an application of the second mask to the second sensor data, a third set of data points from among the second sensor data that corresponds with the surface of the AV at the second time; and
navigating the AV based on the filtered second sensor data to exclude the surface of the AV.

9. The system of claim 8 , wherein the one or more processors are further configured to perform operations comprising:
dynamically identifying the third set of data points from among the second sensor data as self-hit data points instead of a surrounding environment of the AV; and
categorizing the self-hit data points within the second sensor data.

10. The system of claim 8 , wherein the one or more processors are further configured to perform operations comprising:
collecting Light Detection and Ranging (LiDAR) test data for the environment around the AV;
generating the first mask, wherein the first mask is a filter mask based on the geometric model that is created based on the LiDAR test data, and wherein identifying the first set of sensor data that corresponds with the surface of the AV is based on the filter mask.

11. The system of claim 8 , wherein the geometric model is based on a combination of a Computer-aided Design (CAD) model of the AV and the second sensor data.

12. The system of claim 8 , wherein the second sensor data comprises Light Detection and Ranging (LiDAR) data.

13. The system of claim 8 , wherein the one or more processors are further configured to perform operations comprising:
collecting test data for the environment around the AV, and wherein the geometric model of the AV is based on the test data.

14. The system of claim 8 , wherein the geometric model comprises one or more buffers to facilitate use of the geometric model on different vehicle types.

15. A non-transitory computer-readable storage medium comprising instructions stored therein, which when executed by one or more processors, cause the processors to perform operations comprising:
collecting first sensor data for an environment around an autonomous vehicle (AV) at a first time;
as the AV navigates, dynamically creating a geometric model of the AV, wherein the geometric model specifies physical boundaries of the AV in three-dimensional (3D) space captured by the first sensor data;
creating a first mask from the geometric model that distinguishes a first set of sensor data that corresponds with a surface of the AV from a second set of sensor data that corresponds with environmental data;
filtering out, based on an application of the first mask to the first sensor data, data points from among the first set of sensor data that corresponds with the surface of the AV;
collecting second sensor data for the environment around the AV at a second time;
updating the geometric model to create a second mask based on the second sensor data, wherein the second mask is modified in accordance with the second sensor data;
filtering out, based on an application of the second mask to the second sensor data, a third set of data points from among the second sensor data that corresponds with the surface of the AV at the second time; and
navigating the AV based on the filtered second sensor data to exclude the surface of the AV.

16. The non-transitory computer-readable storage medium of claim 15 , wherein the one or more processors are further configured to perform operations comprising:
dynamically identifying the third set of data points from among the second sensor data as self-hit data points instead of a surrounding environment of the

Assignees

• Gm Cruise Holdings Llc

Inventors

• Gangundi Sandeep
• Nair Avinash
• Elogeel Abdelrahman
• Vogt Kyle
• Lekkerkerker Cornelis Jacob

Classifications

• G01S7/4876 Primary

  by removing unwanted signals (G01S7/495 takes precedence) · CPC title

• G01S7/4802

  using analysis of echo signal for target characterisation; Target signature; Target cross-section · CPC title

• G01S7/493

  Extracting wanted echo signals · CPC title

• G01S17/931

  of land vehicles · CPC title

• G06F30/15

  Vehicle, aircraft or watercraft design · CPC title