Vehicle localization based on neural network

Patent metadata

Field	Value
Publication number	US-11079497-B2
Application number	US-201816139433-A
Country	US
Kind code	B2
Filing date	Sep 24, 2018
Priority date	Sep 25, 2017
Publication date	Aug 3, 2021
Grant date	Aug 3, 2021

Abstract

A method for determining a location of a vehicle is provided. The method includes receiving, at data processing hardware, a first set of vehicle system data from one or more vehicles. The method also includes determining, at the data processing hardware, a data model based on the first set of vehicle system data. Additionally, the method includes receiving, at the data processing hardware, a second set of vehicle system data associated with the vehicle. The vehicle being different than the one or more vehicles. The method includes determining, using the data processing hardware, a vehicle location based on the second set of vehicle system data and the data model. The method includes displaying, on a user interface of the vehicle in communication with the data processing hardware, the vehicle location.

First claim

What is claimed is:

1. A method for determining a location of a vehicle, the method comprising:
providing a data processing hardware, the data processing hardware comprising a trained neural network having a plurality of layers of nonlinear processing units, and memory hardware in communication with the data processing hardware, the memory hardware storing instructions of a processing algorithm;
executing, by the nonlinear processing units in a processing phase, the instructions of the processing algorithm to cause the nonlinear processing units to perform:
receiving, at the data processing hardware, a current vehicle system data associated with the vehicle, the vehicle being different than the training vehicles vehicle, the current set of vehicle system data including GPS location data, vehicle dynamics and image data generated by sensors or cameras mounted to the vehicle;
receiving a trained neural network model, the trained neural network model being part of the trained neural network and being based upon a training set of vehicle data comprising GPS location data, vehicle dynamics and sensor data associated with one or more vehicles during a training phase, the trained neural network model comprising weights and biases associated with the training set of vehicle data;
determining, using the data processing hardware, a vehicle location based on the current set of vehicle system data and the trained neural network model; and
transmitting, from the data processing hardware to a user interface of the vehicle in communication with the data processing hardware, a command to display the vehicle location.

2. The method of claim 1 , wherein the data processing hardware is supported by the vehicle.

3. The method of claim 1 , wherein the data processing hardware is in communication with the vehicle by way of a shared network.

4. The method of claim 1 , wherein the data processing hardware includes a first data processing hardware supported by the vehicle and a second data processing hardware in communication with the first data processing hardware by way of a shared network.

5. The method of claim 1 , wherein the training set of vehicle system data is associated with one or more vehicles captured during a predetermined period of time.

6. The method of claim 1 , wherein the current set of vehicle system data is received in real-time while the vehicle is maneuvering along a road.

7. A system for determining a location of a vehicle, the system comprising:
data processing hardware in communication with the user display, the data processing hardware comprising a trained neural network having a plurality of layers of nonlinear processing units;
memory hardware in communication with the data processing hardware, the memory hardware storing instructions of a processing algorithm that when executed by the nonlinear processing units in a processing phase cause the nonlinear processing units to execute a method, the method including:
receiving a current set of vehicle system data associated with the vehicle, the current set of vehicle system data including GPS location data, vehicle dynamics and image data generated by sensors or cameras mounted to the vehicle;
receiving a trained neural network model, the trained neural network model being part of the trained neural network and being based upon a training set of vehicle data comprising GPS location data, vehicle dynamics and sensor data associated with one or more vehicles during a training phase, the trained neural network model comprising weights and biases associated with the training set of vehicle data;
determining a vehicle location based on the current set of vehicle system data and the trained neural network model; and

transmitting to a user interface of the vehicle a command to display the vehicle location.

8. The system of claim 7 , wherein the data processing hardware is supported by the vehicle.

9. The system of claim 7 , wherein the data processing hardware is in communication with the vehicle by way of a shared network.

10. The system of claim 7 , wherein the data processing hardware includes a first data processing hardware supported by the vehicle and a second data processing hardware in communication with the first data processing hardware by way of a shared network.

11. The system of claim 7 , wherein the training set of vehicle system data is associated with training vehicles captured during a predetermined period of time.

12. The system of claim 7 , wherein the current set of vehicle system data is received in real-time while the vehicle is maneuvering along a road.

Assignees

• Continental automotive systems inc

Inventors

• Li Dalong
• Froehlich Dominik
• Muharemovic Ibro

Classifications

• G06N3/08

  Learning methods · CPC title

• G06N3/09

  Supervised learning · CPC title

• G06N3/0464

  Convolutional networks [CNN, ConvNet] · CPC title

• G01S19/49 Primary

  whereby the further system is an inertial position system, e.g. loosely-coupled · CPC title

• G01S19/485

  whereby the further system is an optical system or imaging system · CPC title