Control of autonomous vehicle based on determined yaw parameter(s) of additional vehicle

Patent metadata

Field	Value
Publication number	US-10906536-B2
Application number	US-201816173660-A
Country	US
Kind code	B2
Filing date	Oct 29, 2018
Priority date	Apr 11, 2018
Publication date	Feb 2, 2021
Grant date	Feb 2, 2021

Abstract

Determining yaw parameter(s) (e.g., at least one yaw rate) of an additional vehicle that is in addition to a vehicle being autonomously controlled, and adapting autonomous control of the vehicle based on the determined yaw parameter(s) of the additional vehicle. For example, autonomous steering, acceleration, and/or deceleration of the vehicle can be adapted based on a determined yaw rate of the additional vehicle. In many implementations, the yaw parameter(s) of the additional vehicle 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) component of a vehicle, a group of LIDAR data points of a sensing cycle of the phase coherent LIDAR component,
each of the LIDAR data points of the group indicating a corresponding instantaneous range and a corresponding instantaneous velocity for a corresponding point in an environment of the vehicle, and each of the LIDAR data points of the group being generated based on a corresponding sensing event of the phase coherent LIDAR component during the sensing cycle;

determining that a subgroup, of the LIDAR data points of the group, corresponds to an additional vehicle that is in addition to the vehicle;
based on determining that the subgroup corresponds to the additional vehicle:
determining an instantaneous yaw parameter of the additional vehicle based on a plurality of the LIDAR data points of the subgroup; and

adapting autonomous control of the vehicle based on the determined instantaneous yaw parameter of the additional vehicle.

2. The method of claim 1 , wherein determining that the subgroup of the LIDAR data points of the group corresponds to the additional vehicle comprises:
processing the LIDAR data points of the group using a trained neural network model;
generating, based on processing of the LIDAR data points of the group using the trained neural network model, an output that indicates that the subgroup has a vehicle classification; and
determining that the subgroup corresponds to the additional vehicle based on the output indicating that the subgroup has the vehicle classification.

3. The method of claim 1 , wherein determining the instantaneous yaw parameter of the additional vehicle based on the plurality of the LIDAR data points of the subgroup comprises:
determining a first set of one or more of the plurality of LIDAR data points of the subgroup;
determining a second set of one or more of the plurality of LIDAR data points of the subgroup based on the second set being spatially offset from the first set; and
determining the instantaneous yaw parameter based on comparison of the first set to the second set.

4. The method of claim 3 , wherein determining that the first set and the second set are spatially offset relative to one another is based at least in part on the corresponding ranges indicated by the first set and the second set.

5. The method of claim 3 , wherein determining the instantaneous yaw parameter of the additional vehicle based on comparison of the first set to the second set comprises:
determining the instantaneous yaw parameter based on comparison of a first set velocity magnitude of the first set and a second set velocity magnitude of the second set, wherein the first set velocity magnitude is based on the corresponding instantaneous velocities for the first set, and wherein the second set velocity magnitude is based on the corresponding instantaneous velocities for the second set.

6. The method of claim 5 , wherein the instantaneous yaw parameter of the additional vehicle includes a yaw rate of the additional vehicle, and wherein determining the instantaneous yaw parameter of the additional vehicle based on comparison of the first set to the second set further comprises:
determining a distance based on the spatial offset between the first set and the second set; and
determining the instantaneous yaw rate based on:
the comparison of the first set velocity magnitude and the second set velocity magnitude, and
the distance.

7. The method of claim 6 , wherein determining the instantaneous yaw rate based on the comparison of the first set velocity magnitude and the second set velocity magnitude, and the distance comprises:
determining a velocity differential between the first set velocity magnitude and the second set velocity magnitude based on the comparison of the first set velocity magnitude and the second set velocity magnitude; and
converting the velocity differential to the instantaneous yaw rate based on the distance.

8. The method of claim 5 , wherein the instantaneous yaw parameter of the additional vehicle is a lower bound yaw rate of the additional vehicle, and further comprising determining an additional instantaneous yaw parameter of the additional vehicle, that is an upper bound yaw rate, based on comparison of the first set to the second set.

9. The method of claim 8 , further comprising:
determining a distance based on the spatial offset between the first set and the second set;
determining a velocity differential between the first set velocity magnitude and the second set velocity magnitude based on the comparison of the first set velocity magnitude and the second set velocity magnitude;
wherein determining the lower bound yaw rate comprises:
dividing the velocity differential by the distance.

10. The method of claim 9 , wherein determining the upper bound yaw rate comprises:
dividing the velocity differential by: a value that is based on the distance, but reduced in magnitude relative to the distance.

11. The method of claim 5 , wherein the first set includes multiple of the plurality of LIDAR data points and the first set velocity magnitude is determined based on the corresponding instantaneous velocities for the multiple of the plurality of LIDAR data points.

12. The method of claim 11 , wherein the first set velocity magnitude is based on an average of the corresponding instantaneous velocities for the multiple of the plurality of LIDAR data points.

13. The method of claim 3 , wherein the instantaneous yaw parameter includes a yaw rate, and wherein determining the instantaneous yaw parameter of the additional vehicle based on comparison of the first set to the second set comprises:
determining a velocity differential based on comparison of a first set velocity magnitude and a second set velocity magnitude, wherein the first set velocity magnitude is based on the corresponding velocities for the first set, and wherein the second set velocity magnitude is based on the corresponding velocities for the second set;
identifying a stored model for the additional vehicle, the stored model describing geometric features of the additional vehicle; and
converting the velocity differential to the instantaneous yaw rate based on the stored model for the additional vehicle.

14. The method of claim 13 , wherein identifying the stored model for the additional vehicle comprises:
selecting the stored model, from a plurality of candidate stored models, based on the LIDAR data points of the subgroup.

15. The method of claim 1 , wherein the instantaneous yaw parameter includes a yaw rate and a yaw direction, and wherein determining the instantaneous yaw parameter of the additional vehicle based on the plurality of the LIDAR data points of the subgroup comprises:
processing the plurality of the LIDAR data points of the subgroup using a trained neural network model;
generating, based on processing of the LIDAR data points of the group using the trained neural network model, an output that indicates the instantaneous yaw rate and the yaw direction; and
determining the instantaneous yaw rate and the yaw direction based on the output.

16. The method of claim 15 , wherein processing the plurality of the LIDAR data points of the subgroup using the trained neural network model comprises processing all of the LIDAR data points of the subgroup using the trained neural network model.

17. The method of claim 1 , wherein the instantaneous yaw parameter of the additional vehicle is a velocity differential

Assignees

• Aurora Innovation Inc

Inventors

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

Classifications

• G06F18/2178

  based on feedback of a supervisor · CPC title

• G06N3/045

  Combinations of networks · CPC title

• G06F18/214

  Generating training patterns; Bootstrap methods, e.g. bagging or boosting · CPC title

• G06N3/09

  Supervised learning · CPC title

• G06N3/0464

  Convolutional networks [CNN, ConvNet] · CPC title