CHARACTERIZATION OF SENSOR DATA FOR VEHICLE TELEMATICS

• US 20160132618A1
• Filed: 11/12/2014
• Published: 05/12/2016
• Est. Priority Date: 11/12/2014
• Status: Active Grant

• Current Assignee: Mojio Inc.
• Original Assignee: Mojio Inc.
• Inventors: LOVELL, ROBB E.
• Granted Patent: US 10,006,782 B2
• Time in Patent Office: Days
• US Class Current: 1/1
• CPC Class Codes: G01C 21/26 specially adapted for navigation in a road network
• CPC Class Codes: G01D 3/00 Indicating or recording apparatus with provision for the special purposes referred to in the subgroups
• CPC Class Codes: G01D 3/02 with provision for altering or correcting the law of variation

Abstract

A computing system located on-board a vehicle processes the input sensor data to obtain a downsampled representation of one or more time-series measurements. The computing system further identifies one or more characterization parameters to be reported as supplemental information along with the downsampled representation. The computing system processes the input sensor data to obtain a characterization value for each characterization parameter identified based on measurement conditions and communicates the characterizations based on a timing profile. The computing system formats the one or more characterization values to be reported along with the downsampled representation into one or more report messages. The computing system transmits the one or more report messages over a wireless wide area network directed to a server system. The one or more characterization values may be used by the server system as supplemental information in reconstructing the input sensor data from the downsampled representation.

Claims

1. A method performed by a computing system located on-board a vehicle, the method comprising:

receiving input sensor data representing one or more time-series measurements output by one or more sensors located on-board the vehicle;

classifying the one or more time-series measurements into a measurement type of a plurality of supported measurement types;

processing the input sensor data to obtain a downsampled representation of the one or more time-series measurements, the downsampled representation having a downsampled sampling rate selected based on the measurement type;

identifying one or more characterization parameters based on the measurement type to be reported as supplemental information along with the downsampled representation by referencing a database defining an association between a plurality of supported characterization parameters and the plurality of supported measurement types;

processing the input sensor data to obtain a characterization value for each characterization parameter identified;

formatting the one or more characterization values to be reported along with the downsampled representation into one or more report messages; and

transmitting the one or more report messages over a wireless wide area network directed to a server system, the one or more characterization values for use by the server system as supplemental information in reconstructing the input sensor data from the downsampled representation.

2. The method of claim 1, wherein the one or more characterization parameters include one or more of:

(1) an instantaneous measurement of the input sensor data, (2) a minimum measurement of the input sensor data, (3) a maximum measurement of the input sensor data, (4) an average measurement of the input sensor data, (5) a median measurement of the input sensor data, (6) a change measurement of the input sensor data from one or more prior measurements of input sensor data, (7) a quantity measurement of occurrences of the input data satisfying a condition, (8) a duration measurement of the input data satisfying a condition, (9) a quantity measurement of changes of the input data between two or more states, (10) an equation representation of the input data, and/or (11) a native sampling rate of the input data.

3. The method of claim 1, wherein classifying the one or more time-series measurements into the measurement type includes classifying a single time-series measurement into the measurement type;

and

wherein the method further includes classifying one or more additional time-series measurements into an additional measurement type of the plurality of supported measurement types.

4. The method of claim 1, wherein classifying the one or more time-series measurements into the measurement type further includes classifying, into the measurement type, two or more time-series measurements of the input sensor data from two or more sensor devices.

5. The method of claim 4, further comprising:

processing the two or more time-series measurements output from the two or more sensor devices to obtain a combined time-series measurement for the measurement type; and

wherein processing the input sensor data to obtain the downsampled representation of the one or more time-series measurements further includes processing the combined time-series measurement to obtain the downsampled representation of the combined time-series measurement.

6. The method of claim 5, wherein processing the input sensor data to obtain the characterization value for each characterization parameter identified further includes processing the combined time-series measurement to obtain the characterization value for each characterization parameter identified.

7. The method of claim 5, wherein each of the two or more time-series measurements is associated with one or more corresponding characterization parameters such that a corresponding characterization value is obtained for each of the one or more corresponding characterization parameters.

8. The method of claim 1, further comprising:

comparing each characterization value for each characterization parameter to a corresponding condition associated with that characterization parameter; and

wherein formatting the one or more characterization values to be reported along with the downsampled representation into one or more report messages is selectively performed for each data characterization upon satisfaction of the corresponding condition.

9. The method of claim 8, further comprising:

identifying the corresponding condition for each data characterization parameter by referencing the database further defining an association between the plurality of supported characterization parameters and the corresponding condition.

10. The method of claim 1, wherein the computing system located on-board the vehicle includes a telematics control unit that communicates with a on-board vehicle control system of the vehicle via a wired or wireless communications link;

and

wherein receiving the input sensor data further includes receiving the input sensor data from the on-board vehicle control system over the wired or wireless communications link.

11. The method of claim 1, further comprising:

formatting contextual information into the one or more report messages;

wherein the contextual information includes one or more of;

(1) an identifier of the vehicle, (2) an identifier of the computing system located on-board the device, (3) an identifier of a corresponding time of the one or more characterization values, (4) an identifier of a corresponding time period of the one or more characterization values, (5) an identifier of the sampling rate of the downsampled representation, (6) an identifier of a sampling rate of the input sensor data, (7) an identifier of a corresponding time of the downsampled representation and/or input sensor data, (8) an identifier of a corresponding time period of the downsampled representation and/or input sensor data, (9) an identifier of the measurement type classified from the input sensor data; and

/or (10) an identifier of a geographic location of the vehicle at a time that the input sensor data is received.

12. The method of claim 11, wherein formatting the contextual information into the one or more report messages includes inserting the contextual information into a header of the one or more report messages.

13. The method of claim 12, wherein each of the one or more report messages further include a body in addition to the header;

and

wherein formatting the one or more characterization values into the one or more report messages further includes inserting the one or more characterization values into the body of the one or more report messages.

14. The method of claim 12, wherein formatting the one or more characterization values into the one or more report messages further includes inserting the one or more characterization values into the header of the one or more report messages.

15. The method of claim 12, wherein the downsampled representation is reported along with the downsampled representation by formatting the downsampled representation into a body of the one or more report messages.

16. The method of claim 15, wherein the downsampled representation forms an array of sample values within the body of the one or more report messages.

17. The method of claim 1, wherein the one or more report messages form one or more data packets transmitted over the wireless wide area network;

wherein the one or more report messages are among a sequence of a plurality of report messages transmitted by the computing system over time; and

wherein the sequence of the plurality of report messages corresponds to a sequence of time-series measurements output by the one or more sensors located on-board the vehicle over time.

18. The method of claim 1, wherein processing the input sensor data to obtain the characterization value and/or wherein transmitting the characterization value in the one or more reporting messages is performed responsive to a corresponding timing profile for that characterization parameter;

wherein the timing profile defines one or more state changes in the one or more time-series measurements of the measurement type to be satisfied for the characterization value of the characterization parameter to be included in the one or more reporting messages;

wherein the state changes defined by the timing profile including one or more of a level, a leading edge, a trailing edge, a high level, and/or a low level of the one or more time-series measurements.

19. A computing device located on-board a vehicle and physically interfacing with an on-board vehicle control system of the vehicle via a wired communications link, the computing device comprising:

a vehicle interface to interface with the vehicle control system via the wired communications link;

a communications subsystem to communicate over a wireless wide area nework;

a logic subsystem to execute instructions;

a storage subsystem having instructions stored thereon executable by the logic subsystem to;

receive input sensor data representing one or more time-series measurements output by one or more sensors located on-board the vehicle;

classify the one or more time-series measurements into a measurement type of a plurality of supported measurement types;

process the input sensor data to obtain a downsampled representation of the one or more time-series measurements, the downsampled representation having a downsampled sampling rate selected based on the measurement type;

identify one or more characterization parameters based on the measurement type to be reported as supplemental information along with the downsampled representation by referencing a database defining an association between a plurality of supported characterization parameters and the plurality of supported measurement types;

process the input sensor data to obtain a characterization value for each characterization parameter identified;

format the one or more characterization values to be reported along with the downsampled representation into one or more report messages; and

transmit the one or more report messages over the wireless wide area network directed to a server system, the one or more characterization values for use by the server system as supplemental information in reconstructing the input sensor data from the downsampled representation.

20. A method for a server system including one or more server devices, the method comprising:

receiving one or more report messages transmitted over a wireless wide area network by a computing system located on-board a vehicle, the one or more report messages including a downsampled representation of input sensor data that was output by one or more sensors located on-board the vehicle, the one or more report messages further including one or more data characterization values obtained by the computing system located on-board the vehicle from the input sensor data;

processing the one or more report messages to obtain reconstructed input sensor data based on the downsampled representation using the one or more data characterization values;

identifying one or more qualitative linguistic values based on the reconstructed input sensor data by referencing a database of supported qualitative linguistic values having associated conditions to which the reconstructed input sensor data is compared;

storing the one or more qualitative linguistic values and/or the reconstructed input sensor data as profile information in a profile associated with an identifier included in the one or more report messages;

receiving a request to access the profile information from a client device over a wide area network;

authenticating the client device for the profile information requested by the client device; and

transmitting the profile information requested by the client device over the wide area network following authentication of the client device.