Control of autonomous vehicle based on environmental object classification determined using phase coherent LIDAR data

Patent metadata

Field	Value
Publication number	US-11933902-B2
Application number	US-202218092118-A
Country	US
Kind code	B2
Filing date	Dec 30, 2022
Priority date	Apr 11, 2018
Publication date	Mar 19, 2024
Grant date	Mar 19, 2024

Abstract

Determining classification(s) for object(s) in an environment of autonomous vehicle, and controlling the vehicle based on the determined classification(s). For example, autonomous steering, acceleration, and/or deceleration of the vehicle can be controlled based on determined pose(s) and/or classification(s) for objects in the environment. The control can be based on the pose(s) and/or classification(s) directly, and/or based on movement parameter(s), for the object(s), determined based on the pose(s) and/or classification(s). In many implementations, pose(s) and/or classification(s) of environmental object(s) are determined based on data from a phase coherent Light Detection and Ranging (LIDAR) component of the vehicle, such as a phase coherent LIDAR monopulse component and/or a frequency-modulated continuous wave (FMCW) LIDAR component.

First claim

What is claimed is:

1. A method, comprising:
receiving, from a phase coherent Light Detection and Ranging (LIDAR) monopulse component of a vehicle, the phase coherent LIDAR monopulse component being a frequency-modulated (FM) LIDAR monopulse component, a group of FMLIDAR data points collectively capturing a plurality of points in an area of an environment of the vehicle,
each of the FM LIDAR data points of the group indicating an instantaneous corresponding range and an instantaneous corresponding velocity for a corresponding one of the points in the environment, and each being generated based on a corresponding sensing event of the phase coherent FM LIDAR monopulse component,
wherein the corresponding sensing events of the phase coherent FM LIDAR monopulse component each comprise a first receiver sensing event at a first receiver of the phase coherent FM LIDAR monopulse component;

determining that a subgroup, of the FM LIDAR data points of the group, corresponds to an object of a particular classification based at least upon the sensing events and adapting autonomous control of the vehicle based on determining that the subgroup corresponds to the object of the particular classification.

2. The method of claim 1 , further comprising:
determining at least one instantaneous velocity of the object based on at least one of the corresponding velocities of the subgroup, wherein the at least one of the corresponding velocities of the subgroup is utilized in determining the at least one instantaneous velocity of the object based on the determining that the subgroup corresponds to the object;
wherein adapting autonomous control of the vehicle is further based on the at least one instantaneous velocity.

3. The method of claim 2 , wherein determining the at least one instantaneous velocity based on the at least one of the corresponding velocities of the subgroup comprises determining the at least one instantaneous velocity based on a function of a plurality of the corresponding velocities of the subgroup.

4. The method of claim 2 , wherein adapting autonomous control of the vehicle based on the at least one instantaneous velocity comprises:
determining at least one candidate trajectory of the object based on the at least one instantaneous velocity; and
adapting autonomous control of the vehicle based on the at least one candidate trajectory.

5. The method of claim 4 , wherein determining the at least one candidate trajectory of the object based on the at least one instantaneous velocity comprises:
adapting a previously determined velocity, for the object, based on the at least one instantaneous velocity; and
determining the at least one candidate trajectory of the object based on the adaptation of the previously determined velocity.

6. The method of claim 2 , wherein determining the at least one instantaneous velocity based on the at least one of the corresponding velocities of the subgroup comprises:
selecting an instantaneous velocity determination technique, from a plurality of candidate instantaneous velocity determination techniques, based on the instantaneous velocity determination technique being assigned to the particular classification; and
using the selected instantaneous velocity determination technique in determining the at least one instantaneous velocity.

7. The method of claim 2 , wherein determining the at least one instantaneous velocity based on the at least one of the corresponding velocities of the subgroup comprises:
selecting an instantaneous velocity determination technique, from a plurality of candidate instantaneous velocity determination techniques, based on a quantity of the corresponding velocities of the subgroup; and
using the selected instantaneous velocity determination technique in determining the at least one instantaneous velocity.

8. The method of claim 1 , wherein a corresponding angle is defined for each of the FM LIDAR data points of the group, and wherein the corresponding angles are each defined to a degree that is less than a beam-width of corresponding optical output, from a laser of the phase coherent FM LIDAR monopulse component, that resulted in the corresponding sensing event based on which the FM LIDAR data point is generated.

9. The method of claim 1 , further comprising generating a corresponding angle for each of the FM LIDAR data points of the group, wherein generating a given corresponding angle, of the corresponding angles, for a given FM LIDAR data point of the FM LIDAR data points comprises:
generating the given corresponding angle based on comparing:
first data from the first receiver sensing event for the corresponding sensing event of the given FM LIDAR data point, and
second data from a second receiver sensing event for the corresponding sensing event received at a second receiver of the phase coherent FM LIDAR monopulse component of the given FM LIDAR data point.

10. The method of claim 9 , wherein comparing the first data and the second data comprises comparing one or both of:
phase differences between the first data and the second data, and
amplitude differences between the first data and the second data.

11. The method of claim 10 , wherein the first data and the second data are each a respective time varying intermediate frequency waveform, or a respective range-Doppler image.

12. The method of claim 1 , wherein each of the FM LIDAR data points of the group is a corresponding super-resolution FM LIDAR data point constructed based on combining first data from the first receiver sensing event of the corresponding sensing event and second data from a second receiver sensing event at a second receiver of the phase coherent FM LIDAR monopulse component for the corresponding sensing event.

13. The method of claim 1 , wherein each of the FM LIDAR data points of the group further indicate a corresponding intensity, and wherein determining that the subgroup of the FM LIDAR data points of the group corresponds to the object of the particular classification is further based on the corresponding intensities for multiple of the FM LIDAR data points of the subgroup.

14. The method of claim 1 , further comprising:
receiving, from the phase coherent FM LIDAR monopulse component immediately prior to receiving the group, a prior group of prior FM LIDAR data points that collectively capture a plurality of the points in the area of the environment of the vehicle,
each of the prior FM LIDAR data points of the prior group indicating a corresponding prior range and a corresponding prior velocity for the corresponding points in the environment, and each being generated based on a prior corresponding sensing event of the phase coherent FM LIDAR monopulse component;
wherein determining that the subgroup, of the FM LIDAR data points of the group, corresponds to the object of the particular classification is further based on both:
the corresponding prior ranges for multiple of the prior FM LIDAR data points, and
the corresponding prior velocities for multiple of the prior FM LIDAR data points.

15. The method of claim 1 , wherein determining that the subgroup, of the FM LIDAR data points of the group, corresponds to the object of the particular classification comprises:
processing the FM LIDAR data points of the group using a trained machine learning model;
generating, based on processing of the FM LIDAR data points of the group using the trained machine learning model, an output that indicates that the subgroup has the particular classification; and
determining that the subgroup corresponds to the object having the particular classification based on the ou

Assignees

• Aurora Operations Inc

Inventors

• Smith Warren
• Eade Ethan
• Anderson Sterling J
• Bagnell James Andrew
• Nabbe Bartholomeus C
• Urmson Christopher Paul

Classifications

• B60W2754/00

  Output or target parameters relating to objects · CPC title

• B60W2554/4042

  Longitudinal speed · CPC title

• G05D2111/17

  Coherent light, e.g. laser signals · CPC title

• G05D1/43

  Control of position or course in two dimensions [2D] · CPC title

• G05D1/247

  using signals provided by artificial sources external to the vehicle, e.g. navigation beacons · CPC title