Transferring synthetic lidar system data to real world domain for autonomous vehicle training applications

Patent metadata

Field	Value
Publication number	US-2020326717-A1
Application number	US-201916380180-A
Country	US
Kind code	A1
Filing date	Apr 10, 2019
Priority date	Apr 10, 2019
Publication date	Oct 15, 2020
Grant date	—

Abstract

Methods and systems are disclosed for correlating synthetic LiDAR data to a real-world domain for use in training an autonomous vehicle in how to operate in an environment. To do this, the system will obtain a data set of synthetic LiDAR data, transfer the synthetic LiDAR data to a two-dimensional representation, use the two-dimensional representation to train a model of a real-world environment, and use the trained model of the real-world environment to train an autonomous vehicle.

First claim

1 . A method of correlating synthetic LiDAR data to a real-world domain for training an autonomous vehicle, the method comprising, by a processing device:
obtaining a data set of synthetic LiDAR data; transferring the synthetic LiDAR data to a two-dimensional representation; using the two-dimensional representation to train a model of a real-world environment; and using the trained model of the real-world environment to train an autonomous vehicle.

2 . The method of claim 1 , wherein obtaining the data set of synthetic LiDAR data comprises using a simulation application to generate the data set of synthetic LiDAR data by operating a virtual vehicle in a simulated environment of the simulation application.

3 . The method of claim 2 , wherein using the simulation application to generate the data set of synthetic LiDAR data comprises, as the virtual vehicle is operated in the simulated environment:
accessing a plurality of depth images generated by the simulation; extracting samples from a plurality of image frames generated by the simulation; and correlating the extracted samples to the depth images to approximate each extracted sample to a pixel in one or more of the depth images.

4 . The method of claim 1 , wherein transferring the synthetic LiDAR data to the two-dimensional representation comprises transferring the synthetic LiDAR data to a birds-eye view representation.

5 . The method of claim 1 , wherein transferring the synthetic LiDAR data to the two-dimensional representation comprises generating a plurality of two-dimensional front-view images.

6 . The method of claim 1 , further comprising generating ground truth annotations in a plurality of frames of synthetic LiDAR data to identify one or more objects in the synthetic LiDAR data.

7 . The method of claim 1 , wherein:
the method further comprises receiving a plurality of images captured from the real-world environment; and transferring the synthetic LiDAR data to the two-dimensional representation comprises:
generating a plurality of two-dimensional front-view images from the LiDAR data, and
mapping one or more of the objects in the synthetic LiDAR data to one or more objects in the plurality of images.

8 . The method of claim 7 , wherein:
the method further comprises generating ground truth annotations in a plurality of frames of the data set of synthetic LiDAR data to identify one or more objects in the synthetic LiDAR data; and wherein at least some of the front-view images comprise one or more of the ground truth annotations that identify one or more objects.

9 . The method of claim 7 , wherein using the two-dimensional representation to train the model of the real-world environment comprises using the plurality of images captured from the real-world environment, the two-dimensional front-view images from the LiDAR data, and a result of the mapping to train the model of the real-world environment.

10 . A system for correlating synthetic LiDAR data to a real-world domain for training an autonomous vehicle, the system comprising:
a processor; a non-transitory memory that stores a data set of synthetic LiDAR data; and a non-transitory memory containing programming instructions that are configured to cause the processor to:
access the data set of synthetic LiDAR data and extract synthetic LiDAR data from the data set,
transfer the extracted synthetic LiDAR data to a two-dimensional representation,
use the two-dimensional representation to train a model of a real-world environment, and
use the trained model of the real-world environment to train an autonomous vehicle.

11 . The system of claim 10 , further comprising additional programming instructions that are configured to cause a processor to implement a simulation application that generates the data set of synthetic LiDAR data by operating a virtual vehicle in a simulated environment of the simulation application.

12 . The system of claim 11 , wherein the programming instructions to implement the simulation application that generates the data set of synthetic LiDAR data comprise instructions to, as the virtual vehicle is operated in the simulated environment:
access a plurality of depth images generated by the simulation; extract samples from a plurality of image frames generated by the simulation; and correlate the extracted samples to the depth images to approximate each extracted sample to a pixel in one or more of the depth images.

13 . The system of claim 10 , wherein the programming instructions to transfer the synthetic LiDAR data to the two-dimensional representation comprises instructions to transfer the synthetic LiDAR data to a birds-eye view representation.

14 . The system of claim 10 , wherein the programming instructions to transfer the synthetic LiDAR data to the two-dimensional representation comprise instructions to generate a plurality of two-dimensional front-view images.

15 . The system of claim 10 , further comprising additional programming instructions that are configured to cause the processor to generate ground truth annotations in a plurality of frames of synthetic LiDAR data to identify one or more objects in the synthetic LiDAR data.

16 . The system of claim 10 :
further comprising additional programming instructions that are configured to cause the system to receive a plurality of images captured from the real-world environment; and wherein the instructions to transfer the synthetic LiDAR data to the two-dimensional representation comprise instructions to:
generate a plurality of two-dimensional front-view images from the LiDAR data, and
map one or more of the objects in the synthetic LiDAR data to one or more objects in the plurality of images.

17 . The system of claim 16 :
further comprising additional programming instructions that are configured to cause the processor to generate ground truth annotations in a plurality of frames of the data set of synthetic LiDAR data to identify one or more objects in the synthetic LiDAR data; and wherein at least some of the front-view images comprise one or more of the ground truth annotations that identify one or more objects.

18 . The system of claim 16 , wherein the instructions to use the two-dimensional representation to train the model of the real-world environment comprise instructions to use the plurality of images captured from the real-world environment, the two-dimensional front-view images from the LiDAR data, and a result of the mapping to train the model of the real-world environment.

19 . A non-transitory computer-readable medium containing programming instructions for correlating synthetic LiDAR data to a real-world domain and training an autonomous vehicle, the instructions comprising instructions to:
receive a plurality of images captured from a real-world environment; implement a simulation application that generates a data set of synthetic LiDAR data by operating a virtual vehicle in a simulated environment of the simulation application; extract synthetic LiDAR data from the data set; transfer the extracted synthetic LiDAR data to a two-dimensional representation by:
generating a plurality of two-dimensional front-view images from the LiDAR data, and
mapping one or more of the objects in the synthetic LiDAR data to one or more objects in the plurality of images;
use the two-dimensional representation to train a model of the real-world environment; and use the trained model of the real-world environment to train an autonomous vehicle.

20 . T

Assignees

• Argo Ai Llc

Inventors

• Chen Kevin
• Hays James
• Yumer Ersin

Classifications

• G06V20/56 Primary

  exterior to a vehicle by using sensors mounted on the vehicle · CPC title

• G06V10/774

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

• G06V10/803

  of input or preprocessed data · CPC title

• G06V10/82

  using neural networks · CPC title

• G06F18/251

  of input or preprocessed data · CPC title