Method of Programming a Neural Network Computer

• US 20150161505A1
• Filed: 12/11/2013
• Published: 06/11/2015
• Est. Priority Date: 12/11/2013
• Status: Active Grant

• Current Assignee: Volvo Car Corporation (AB Volvo)
• Original Assignee: Volvo Car Corporation (AB Volvo)
• Inventors: Sugioka, Ichiro
• Granted Patent: US 9,361,575 B2
• Time in Patent Office: Days
• US Class Current: 1/1
• CPC Class Codes: G06N 3/02 Neural networks
• CPC Class Codes: G06V 20/59 inside of a vehicle, e.g. relating to seat occupancy, driver state or inner lighting conditions

Abstract

A method is disclosed for programming a target neural network computer for use in cognitive computing systems in automotive safety applications. An observer neural network computer is integrated into active safety systems of a plurality of vehicles to observe signals. Each respective observer neural network computer is arranged to observe signals from a forward facing camera and signals from a driver action monitor of its respective vehicle, to process the observed signals from the forward facing camera of its respective vehicle and correlate them with the observed signals from the driver action monitor of its respective vehicle. The correlated signals from the plurality of observer neural network computers are combined, and the target neural network computer is programmed for use in cognitive computing systems in automotive safety applications based on said combined correlated signals.

Claims

1. A method of programming a target neural network computer for use in cognitive computing systems in automotive safety applications, the method comprising:

integrating a plurality of observer neural network computers into active safety systems of a plurality of vehicles, each observer neural network integrated into one of the plurality of vehicles, the observer neural network computers integrated to observe signals;

arranging each observer neural network computer to observe signals from a forward facing camera of its respective vehicle;

arranging each observer neural network computer to observe signals from one or more driver action monitoring devices of its respective vehicle;

arranging each observer neural network computer to process the observed signals from the forward facing camera of its respective vehicle and to correlate them with the observed signals from the one or more driver action monitoring devices of its respective vehicle;

combining correlated signals from the plurality of observer neural network computers; and

programming the target neural network computer for use in cognitive computing systems in automotive safety applications based on the combined correlated signals.

2. The method of claim 1 wherein arranging each observer neural network computer to observe signals from the forward facing camera of its respective vehicle comprises arranging each observer neural network computer to observe signals from a forward facing pedestrian detection camera of its respective vehicle.

3. The method of claim 1 wherein arranging each observer neural network computer to observe signals from one or more driver action monitoring devices of its respective vehicle comprises arranging each observer neural network computer to observe signals from at least one driver monitoring camera.

4. The method of claim 1 wherein arranging each observer neural network computer to process the observed signals from the forward facing camera of its respective vehicle comprises arranging each observer neural network computer to process the observed signals from the forward facing camera of its respective vehicle to recognize visual patterns and to correlate said visual patterns with the observed signals from the one or more driver action monitoring devices of its respective vehicle.

5. The method of claim 1 wherein arranging each observer neural network computer to observe signals from one or more driver action monitoring devices of its respective vehicle comprises arranging each observer neural network computer to observe signals from at least one of a steering wheel angle sensor, a throttle position sensor, a brake pedal position sensor, activation of a turn indicator, activation of a hazard warning, activation of headlights and activation of windscreen wipers.

6. The method of claim 5 wherein integrating a plurality of observer neural network computers into active safety systems of a plurality of vehicles comprises integrating each observer neural network into one of the plurality of vehicles only to observe signals and integrating a respective observer neural network computer into respective active safety systems of at least one hundred vehicles.

7. The method of claim 1 wherein combining correlated signals from said plurality of said observer neural network computers comprises combining correlated signals from at least one hundred vehicles.

15. A vehicle having a target neural network for use in a cognitive computing system in an automotive safety application, the target neural network programmed according to the method of claim 1.

17. A target neural network for use in a cognitive computing system in an automotive safety application, the target neural network programmed according to the method of claim 1

8. A method for use in programming a target neural network computer for a cognitive computing system in an automotive safety application, the method comprising:

integrating a plurality of observer neural network computers into active safety systems of a plurality of vehicles, each observer neural network integrated into one of the plurality of vehicles;

arranging each observer neural network computer to observe signals from a forward facing camera of its respective vehicle;

arranging each observer neural network computer to observe signals from one or more driver action monitoring devices of its respective vehicle;

arranging each observer neural network computer to process the observed signals from the forward facing camera of its respective vehicle and to correlate them with the observed signals from the one or more driver action monitoring devices of its respective vehicle; and

combining correlated signals from the plurality of observer neural network computers, the correlated signals for use in programming the target neural network computer.

9. The method of claim 8 wherein arranging each observer neural network computer to observe signals from the forward facing camera of its respective vehicle comprises arranging each observer neural network computer to observe signals from a forward facing pedestrian detection camera of its respective vehicle.

10. The method of claim 8 wherein arranging each observer neural network computer to observe signals from one or more driver action monitoring devices of its respective vehicle comprises arranging each observer neural network computer to observe signals from at least one driver monitoring camera.

11. The method of claim 8 wherein arranging each observer neural network computer to process the observed signals from the forward facing camera of its respective vehicle comprises arranging each observer neural network computer to process the observed signals from the forward facing camera of its respective vehicle to recognize visual patterns and to correlate said visual patterns with the observed signals from the one or more driver action monitoring devices of its respective vehicle.

12. The method of claim 8 wherein arranging each observer neural network computer to observe signals from one or more driver action monitoring devices of its respective vehicle comprises arranging each observer neural network computer to observe signals from at least one of a steering wheel angle sensor, a throttle position sensor, a brake pedal position sensor, activation of a turn indicator, activation of a hazard warning, activation of headlights and activation of windscreen wipers.

13. The method of claim 12 wherein integrating a plurality of observer neural network computers into active safety systems of a plurality of vehicles comprises integrating each observer neural network into one of the plurality of vehicles only to observe signals and integrating a respective observer neural network computer into respective active safety systems of at least one hundred vehicles.

14. The method of claim 8 wherein combining correlated signals from said plurality of said observer neural network computers comprises combining correlated signals from at least one hundred vehicles.

16. A vehicle having a target neural network a cognitive computing system in an automotive safety application, the target neural network programmed according to the method of claim 8.

18. A target neural network for a cognitive computing system in an automotive safety application, the target neural network programmed according to the method of claim 8.

19. A method of programming a target neural network computer for use in cognitive computing systems in automotive safety applications, the method comprising:

combining correlated signals from a plurality of observer neural network computers, the plurality of observer neural network computers integrated in active safety systems of a plurality of vehicles, each observer neural network integrated in one of the plurality of vehicles, each observer neural network computer arranged to observe signals from a forward facing camera of its respective vehicle and to observe signals from one or more driver action monitoring devices of its respective vehicle, each observer neural network computer arranged to process the observed signals from the forward facing camera of its respective vehicle and to correlate them with the observed signals from the one or more driver action monitoring devices of its respective vehicle; and

programming the target neural network computer for use in cognitive computing systems in automotive safety applications based on the combined correlated signals.

20. A target neural network for a cognitive computing system in an automotive safety application, the target neural network programmed according to the method of claim 19.