Identifying and/or removing false positive detections from lidar sensor output

Patent metadata

Field	Value
Publication number	US-11480686-B2
Application number	US-202016851033-A
Country	US
Kind code	B2
Filing date	Apr 16, 2020
Priority date	Mar 27, 2019
Publication date	Oct 25, 2022
Grant date	Oct 25, 2022

Abstract

Particulate matter, such as dust, steam, smoke, rain, etc. may cause one or more sensor types to generate false positive detections. In particular, various depth measurements may be impeded by particulate matter. Identifying a false return and/or removing a false detection based at least in part on a sensor output may comprise determining a similarity of a portion of a return signal to an emitted light pulse or an expected return signal, determining a variance of the signal portion over time, determining a difference between a power spectrum of the return relative to an expected power spectrum, and/or determining that a duration associated with the signal portion meets or exceeds a threshold duration.

First claim

What is claimed is:

1. A method comprising:
receiving, from a sensor of a LIDAR device responsive to light incident on the sensor, a first output signal associated with a first distance;
identifying, as a first return, a first portion of the first output signal that meets or exceeds a detection threshold;
identifying the first portion as a false positive return based at least in part on at least one of:
determining a similarity of the first portion of the first output signal that meets or exceeds a detection threshold to an emitted light pulse or an expected return signal;
determining a variance of the first portion of the first output signal that meets or exceeds the detection threshold over time;
determining a difference between a power spectrum of the first return relative to an expected power spectrum of the expected return signal; or
determining that a duration associated with the first portion of the first output signal that meets or exceeds the detection threshold meets or exceeds a threshold duration; and

outputting, as a false detection, at least one of a first distance measurement, an uncertainty, or an identification of the first portion of the first output signal that meets or exceeds the detection threshold as being a false positive return.

2. The method of claim 1 , further comprising:
receiving a second output signal associated with a second distance; and
identifying, as one or more second returns, one or more portions of the second output signal that meet or exceed the detection threshold.

3. The method of claim 2 , further comprising:
emitting a first light pulse, the first light pulse comprising a first power and first duration; and
emitting a second light pulse comprising a second power and a second duration, wherein at least one of the second power differs from the first power or the second duration differs from the first duration,
wherein the first output signal is associated with the first light pulse and the second output signal is associated with the second light pulse.

4. The method of claim 2 , wherein identifying the first portion as a false positive further comprises at least one of:
determining a spectral representation based at least in part on the first distance and the second distance;
determining at least one of a number of frequencies or a power associated with a frequency represented in the spectral representation;
determining a distribution of distances associated with the LIDAR device relative to a mean of the first distance, the second distance, and one or more additional distances; or
determining that a distance between a first point associated with the first distance and a second point associated with the second distance meets or exceeds a threshold distance.

5. The method of claim 1 , further comprising controlling an autonomous vehicle based at least in part on the first distance measurement.

6. The method of claim 1 , further comprising:
receiving a thermal image,
wherein identifying the first return as the false positive return is further based at least in part on determining that a portion of the thermal image is associated with an outlier temperature.

7. A system comprising:
one or more processors; and
a memory storing processor-executable instructions that, when executed by the one or more processors, cause the system to perform operations comprising:
receiving, from a LIDAR device, a first output signal associated with a first distance;
identifying, as a first return, a first portion of the first output signal;
identifying the first portion as a false positive return based at least in part on at least one of:
determining a similarity of the first portion to an emitted light pulse or an expected return signal;
determining a variance of the first portion over time;
determining a difference between a power spectrum of the first return relative to an expected power spectrum of the expected return signal; or
determining that a duration associated with the first portion meets or exceeds a threshold duration; and

outputting, as a false detection, at least one of a first distance measurement, an uncertainty, or an identification of the first portion as being a false positive return.

8. The system of claim 7 , wherein the operations further comprise:
receiving a second output signal associated with a second distance; and
identifying, as one or more second returns, one or more portions of the second output signal that meet or exceed a detection threshold.

9. The system of claim 8 , wherein the operations further comprise:
emitting a first light pulse, the first light pulse comprising a first power and first duration; and
emitting a second light pulse comprising a second power and a second duration, wherein at least one of the second power differs from the first power or the second duration differs from the first duration,
wherein the first output signal is associated with the first light pulse and the second output signal is associated with the second light pulse.

10. The system of claim 9 , wherein the operations further comprise:
identifying, as a true positive return, a last return of the one or more second returns that is last-in-time;
identifying, as a true positive return, a second return of the one or more second returns that is associated with a second similarity that meets or exceeds a similarity threshold, a second variance that is less than a variance threshold, and/or a third duration that is less than the threshold duration; or
identifying, as a false positive, a third return of the one or more second returns that is associated with third similarity that is less than the similarity threshold, a third variance that is meets or exceeds the variance threshold, and/or a fourth duration that meets or exceeds the threshold duration.

11. The system of claim 9 , wherein identifying the first portion as a false positive further comprises at least one of:
determining a spectral representation based at least in part on the first distance and the second distance;
determining at least one of a number of frequencies or a power associated with a frequency represented in the spectral representation;
determining a distribution of distances associated with the LIDAR device relative to a mean of the first distance, the second distance, and one or more additional distances; or
determining that a distance between a first point associated with the first distance and a second point associated with the second distance meets or exceeds a threshold distance.

12. The system of claim 7 , wherein the operations further comprise:
receiving a thermal image,
wherein identifying the first return as a false positive return is further based at least in part on determining that a portion of the thermal image is associated with an outlier temperature.

13. The system of claim 7 , wherein the threshold duration is based at least in part on a duration associated with a signal transmitted by the LIDAR device.

14. A non-transitory computer-readable medium storing computer-executable instructions that, when executed by one or more processors, cause the one or more processors to perform operations comprising:
receiving, from a LIDAR device, a first output signal associated with a first distance;
identifying, as a first return, a first portion of the first output signal;
identifying the first portion as a false positive return based at least in part on at least one of:
determining a similarity of the first portion to an emitted light pulse or an expected return signal;
determining a variance of the first portion over time;
determining a difference between

Assignees

• Zoox Inc

Inventors

• Bhaskaran Sreevatsan
• Ferdowsi Mehran
• Mcmichael Ryan
• Subasingha Subasingha Shaminda

Classifications

• G01S7/497

  Means for monitoring or calibrating · CPC title

• G01S17/86

  Combinations of lidar systems with systems other than lidar, radar or sonar, e.g. with direction finders · CPC title

• G01S17/931 Primary

  of land vehicles · CPC title

• G01S7/484

  Transmitters · CPC title

• G01S17/42

  Simultaneous measurement of distance and other co-ordinates (indirect measurement G01S17/46) · CPC title