MACHINE LEARNING BASED DETERMINATION OF ACCURATE MOTION PARAMETERS OF A VEHICLE

• US 20170103342A1
• Filed: 10/12/2016
• Published: 04/13/2017
• Est. Priority Date: 10/13/2015
• Status: Abandoned Application

• Current Assignee: Athena Vision, LLC
• Original Assignee: Flywheel Software Incorporated (Cabconnect, Inc.)
• Inventors: Rajani, Purshotam, Nukala, Gaurav R., Kannan, Gokulnath Coimbatore, Essy, Blair R.
• Application Number:
  US15/291,500
• Publication Number:
  US 20170103342A1
• Time in Patent Office: Days
• CPC Class Codes: B60R 16/0231 Circuits relating to the driving or the functioning of the vehicle monitoring tyres B60C23/00; indicating overspeed B60K31/00; for dash boards B60K37/00, B60Q3/10; for indicating emergencies B60Q1/52; brake control systems B60T; registering or indicating the working of vehicles G07C5/00; measuring distance G01C, e.g. combinations of speed and distance G01C23/00; engine indicators G01L; measuring speed or acceleration G01P
• CPC Class Codes: G01C 21/26 specially adapted for navigation in a road network
• CPC Class Codes: G06N 20/00 Machine learning
• CPC Class Codes: G06N 5/04 Inference or reasoning models
• CPC Class Codes: G07B 13/00 Taximeters measuring distance travelled G01C; measuring time G04
• CPC Class Codes: G07C 5/008 communicating information to a remotely located station transmission systems for measured values G08C
• CPC Class Codes: G07C 5/0808 Diagnosing performance data testing of vehicles G01M17/00; testing of electrical installation on vehicles G01R31/005

Abstract

A multi-modal meter of a vehicle obtains information from multiple sources to determine the most accurate values of motion parameters of the vehicle. The multi-modal meter obtains data describing motion of a vehicle from various sources including an on-board diagnostics (OBD) and global positioning system (GPS.) The dynamically evaluates the signal sources for their accuracy as the vehicle travels. The multi-modal meter selects different signal sources for different portions of a ride and uses the data from the selected signal sources to determine the most accurate motion parameters. The multi-modal meter use machine learning techniques to generate metadata used by an engine configured to determine the most accurate values of motion parameters of the vehicle.

Claims

1. An apparatus, comprising:

a processor;

local signal sources comprising one or more of;

a global positioning system (GPS), a motion sensor, an accelerometer, a magnetometer, or a camera;

an interface for communicating with external signal sources; and

a non-transitory computer readable storage medium storing instructions, wherein the instructions cause the processor to;

receive a machine learning model, the machine learning module configured to process ride parameters describing a ride or a portion of the ride of a vehicle, wherein the machine learning model is for determination of accurate motion parameters for the vehicle;

for each of a plurality of portions of the ride;

receive ride parameters from a plurality of signal sources, wherein the ride parameters describe the portion of the ride, wherein each signal source is one of a local signal source or an external signal source;

extract a feature vector from the received ride parameters, the feature vector comprising at least a feature describing the vehicle;

execute the machine learning model by providing the feature vector as input to the machine learning model;

determine a set of accurate motion parameters describing the portion of the ride based on the execution of the machine learning model, wherein the set of accurate motion parameters comprises one or more of;

a speed of the vehicle, time duration of the portion of ride, acceleration of the vehicle, or location of the vehicle; and

determine a weighted aggregate value associated with the ride based on the set of accurate motion parameters for the ride.

2. The apparatus of claim 1, wherein the machine learning model is further configured to generate a score for determining an accuracy threshold for measuring accuracy of a signal source.

3. The apparatus of claim 1, wherein the machine learning model is further configured to generate a score for determining precedence of rules executed for determining the set of accurate motion parameters.

4. The apparatus of claim 1, wherein the machine learning model is received from an external system, wherein the external system trains the machine learning model based on past ride parameters obtained from a plurality of vehicles.

5. The apparatus of claim 1, wherein the feature vector comprises a feature representing one of:

make and model of the vehicle, a make and model of a device of the apparatus, or a version of software executing on the device of the apparatus.

6. The apparatus of claim 1, wherein the feature vector comprises a feature representing one of:

location of the vehicle, acceleration of the vehicle, or speed of the vehicle.

7. The apparatus of claim 1, wherein the feature vector comprises a feature representing one of:

a time of day, or a day of the week.

8. A method for determining accurate motion parameters for a vehicle, the method comprising:

receiving a machine learning model configured to process ride parameters describing a ride or a portion of the ride of a vehicle, wherein the machine learning model is for determination of accurate motion parameters for the vehicle;

for each of a plurality of portions of the ride;

receiving ride parameters from a plurality of signal sources, wherein the ride parameters describe the portion of the ride and each signal source is one of a local signal source or an external signal source;

extracting a feature vector from the received ride parameters, the feature vector comprising at least a feature describing the vehicle;

executing the machine learning model by providing the feature vector as input to the machine learning model;

determining a set of accurate motion parameters describing the portion of the ride based on the execution of the machine learning model, wherein the set of accurate motion parameters comprises one or more of;

a speed of the vehicle, time duration of the portion of ride, acceleration of the vehicle, or location of the vehicle; and

determining a weighted aggregate value associated with the ride based on the set of accurate motion parameters for the ride.

9. The method of claim 8, wherein the machine learning model is configured to generate a score for determining an accuracy threshold for measuring accuracy of a signal source.

10. The method of claim 8, wherein the machine learning model is configured to generate a score for determining precedence of rules executed for determining the set of accurate motion parameters.

11. The method of claim 8, wherein the information describing the machine learning model is received from an external system, wherein the external system trains the machine learning model based on past ride parameters obtained from a plurality of vehicles.

12. The method of claim 8, wherein the feature vector comprises a feature representing one of:

make and model of the vehicle, a make and model of a device of the apparatus, or a version of software executing on the device of the apparatus.

13. The method of claim 8, wherein the feature vector comprises a feature representing one of:

location of the vehicle, acceleration of the vehicle, or speed of the vehicle.

14. The method of claim 8, wherein the feature vector comprises a feature representing one of:

a time of day, or a day of the week.

15. A non-transitory computer readable storage medium storing instructions, wherein the instructions cause a processor to:

receive a machine learning model configured to process ride parameters describing a ride or a portion of the ride of a vehicle, wherein the machine learning model is for determination of accurate motion parameters for the vehicle;

for each of a plurality of portions of the ride;

receive ride parameters from a plurality of signal sources, wherein the ride parameters describe the portion of the ride and each signal source is one of a local signal source or an external signal source;

extract a feature vector from the received ride parameters, the feature vector comprising at least a feature describing the vehicle;

execute the machine learning model by providing the feature vector as input to the machine learning model;

determine a set of accurate motion parameters describing the portion of the ride based on the execution of the machine learning model, wherein the set of accurate motion parameters comprises one or more of;

a speed of the vehicle, time duration of the portion of ride, acceleration of the vehicle, or location of the vehicle; and

determine a weighted aggregate value associated with the ride based on the set of accurate motion parameters for the ride.

16. The non-transitory computer readable storage medium of claim 15, wherein the machine learning model is further configured to generate a score for determining an accuracy threshold for measuring accuracy of a signal source.

17. The non-transitory computer readable storage medium of claim 15, wherein the machine learning model is further configured to generate a score for determining precedence of rules executed for determining the set of accurate motion parameters.

18. The non-transitory computer readable storage medium of claim 15, wherein the information describing the machine learning model is received from an external system, wherein the external system trains the machine learning model based on past ride parameters obtained from a plurality of vehicles.

19. The non-transitory computer readable storage medium of claim 15, wherein the feature vector comprises a feature representing one of:

make and model of the vehicle, a make and model of a device of the apparatus, or a version of software executing on the device of the apparatus.

20. The non-transitory computer readable storage medium of claim 15, wherein the feature vector comprises a feature representing one of:

location of the vehicle, acceleration of the vehicle, speed of the vehicle, a time of day, or a day of the week.