Occupant profiling system

• US 10,112,505 B2
• Filed: 09/21/2016
• Issued: 10/30/2018
• Est. Priority Date: 09/21/2016
• Status: Active Grant

• Current Assignee: Intel Corporation
• Original Assignee: Intel Corporation
• Inventors: Tanriover, Cagri Cagatay, Luebbers, Enno Luebbert Egge, Sargut, Selin
• Primary Examiner(s): Brushaber, Frederick M
• Application Number:
  US15/271,959
• Publication Number:
  US 20180079322A1
• Time in Patent Office: 769 Days
• CPC Class Codes: B60N 2/002 Seats provided with an occupancy detection means mounted therein or thereon detection means responsive to presence or absence of children B60N2/266
• CPC Class Codes: B60N 2/0244 with logic circuits
• CPC Class Codes: B60N 2/0248 with memory of positions
• CPC Class Codes: B60N 2/0268 using sensors or detectors for adapting the seat or seat part, e.g. to the position of an occupant
• CPC Class Codes: B60R 21/01512 Passenger detection systems
• CPC Class Codes: B60R 21/01538 for image processing, e.g. cameras or sensor arrays

Abstract

Various systems and methods for providing an occupant profiling system are described herein. A system for profiling a vehicle occupant includes a vehicle control system, comprising: a sensor fusion circuit coupled to a plurality of sensors installed a vehicle seat, the sensor fusion circuit to: determine that a human occupant is seated in the vehicle seat; sample a backrest sensor arrangement and a base cushion sensor arrangement of the vehicle seat to obtain backrest sensor data and base cushion sensor data, when the human occupant is determined to be seated in the vehicle seat; sample, when the human occupant is determined to be seated in the vehicle seat, a height sensor to obtain a height sensor value; and statistically determine from the backrest sensor data, the base cushion sensor data, and the high sensor data, whether the human occupant matches a reference profile.

Claims

1. A system for profiling a vehicle occupant, the system comprising:

a vehicle control system, comprising;

a sensor fusion circuit coupled to a plurality of sensors installed a vehicle seat, the sensor fusion circuit to;

determine that a human occupant is seated in the vehicle seat;

sample a backrest sensor arrangement and a base cushion sensor arrangement of the vehicle seat to obtain backrest sensor data and base cushion sensor data, when the human occupant is determined to be seated in the vehicle seat;

sample, when the human occupant is determined to be seated in the vehicle seat, a height sensor to obtain height sensor data; and

statistically determine from the backrest sensor data, the base cushion sensor data, and the height sensor data, whether the human occupant matches a reference profile, the reference profile used as a baseline to identify the human occupant.

2. The system of claim 1, wherein to determine that the human occupant is seated in the vehicle seat, the sensor fusion circuit is to:

interface with an image processor, the image processor coupled to a camera array interface and operable to;

obtain an image of the vehicle seat; and

use a facial detection process on the image to determine whether a face is present in the image.

3. The system of claim 2, wherein the image processor is to conditionally use the facial detection process when the base cushion sensor data indicates that more than a threshold weight is on the vehicle seat.

4. The system of claim 1, wherein the height sensor comprises an ultrasonic sensor.

5. The system of claim 1, wherein the height sensor data comprises a distance between the height sensor and an upper portion of the human occupant.

6. The system of claim 1, wherein the reference profile comprises a reference backrest sensor data set, a reference base cushion sensor data set, and a reference height sensor data set, and wherein to statistically determine whether the human occupant matches the reference profile, the sensor fusion circuit is to:

obtain a backrest confidence value indicating the likelihood that the backrest sensor data corresponds to the reference backrest sensor data set;

obtain a base cushion confidence value indicating the likelihood that the base cushion sensor data corresponds to the reference base cushion sensor data set;

obtain a height confidence value indicating the likelihood that the seat height sensor data corresponds to the reference height sensor data set; and

aggregate the backrest confidence value, the base cushion confidence value, and the height confidence value to obtain a total confidence value.

7. The system of claim 6, wherein the backrest sensor data comprises a set of sensor values corresponding to a plurality of sensors in the backrest sensor arrangement.

8. The system of claim 7, wherein to obtain the backrest confidence value, the sensor fusion circuit is to:

calculate a sum of sums of dot products of the set of sensor values corresponding to the plurality of sensors in the backrest sensor arrangement and a set of sensor values corresponding to the reference backrest sensor data set; and

compare the sum of sums with a sum of sums of the set of sensor values corresponding to the reference backrest sensor data set with itself.

9. The system of claim 6, wherein the base cushion sensor data comprises a set of sensor values corresponding to a plurality of sensors in the base cushion sensor arrangement.

10. The system of claim 9, wherein to obtain the base cushion confidence value, the sensor fusion circuit is to:

calculate a sum of sums of dot products of the set of sensor values corresponding to the plurality of sensors in the base cushion sensor arrangement and a set of sensor values corresponding to the reference base cushion sensor data set; and

compare the sum of sums with a sum of sums of the set of sensor values corresponding to the reference base cushion sensor data set with itself.

11. The system of claim 6, wherein to obtain the height confidence value, the sensor fusion circuit is to:

calculate a difference of the height sensor data and the reference height sensor data; and

use the difference to determine the height confidence value.

12. The system of claim 1, wherein to aggregate the backrest, base cushion, and height confidence values, the sensor fusion circuit is to:

aggregate the backrest, base cushion, and height confidence values in a weighted function.

13. A method of profiling a vehicle occupant, the method comprising:

determining that a human occupant is seated in a vehicle seat;

sampling, at a sensor fusion circuit, a backrest sensor arrangement and a base cushion sensor arrangement of the vehicle seat to obtain backrest sensor data and base cushion sensor data, when the human occupant is determined to be seated in the vehicle seat;

sampling, at the sensor fusion circuit when the human occupant is determined to be seated in the vehicle seat, a height sensor to obtain height sensor data; and

statistically determining from the backrest sensor data, the base cushion sensor data, and the height sensor data, whether the human occupant matches a reference profile, the reference profile used as a baseline to identify the human occupant.

14. The method of claim 13, wherein determining that the human occupant is seated in the vehicle seat comprises:

obtaining an image of the vehicle seat; and

using a facial detection process on the image to determine whether a face is present in the image.

15. The method of claim 14, further comprising conditionally using the facial detection process when the base cushion sensor data indicates that more than a threshold weight is on the vehicle seat.

16. The method of claim 13, wherein the height sensor comprises an ultrasonic sensor.

17. The method of claim 13, wherein the height sensor data comprises a distance between the height sensor and an upper portion of the human occupant.

18. The method of claim 13, wherein the reference profile comprises a reference backrest sensor data set, a reference base cushion sensor data set, and a reference height sensor data set, and wherein statistically determining whether the human occupant matches the reference profile comprises:

obtaining a backrest confidence value indicating the likelihood that the backrest sensor data corresponds to the reference backrest sensor data set;

obtaining a base cushion confidence value indicating the likelihood that the base cushion sensor data corresponds to the reference base cushion sensor data set;

obtaining a height confidence value indicating the likelihood that the seat height sensor data corresponds to the reference height sensor data set; and

aggregating the backrest confidence value, the base cushion confidence value, and the height confidence value to obtain a total confidence value.

19. The method of claim 13, wherein aggregating the backrest, base cushion, and height confidence values comprises:

aggregating the backrest, base cushion, and height confidence values in a weighted function.

20. At least one non-transitory machine-readable medium including instructions for profiling a vehicle occupant, which when executed by a machine, cause the machine to:

determine that a human occupant is seated in a vehicle seat;

sample, at a sensor fusion circuit, a backrest sensor arrangement and a base cushion sensor arrangement of the vehicle seat to obtain backrest sensor data and base cushion sensor data, when the human occupant is determined to be seated in the vehicle seat;

sample, at the sensor fusion circuit when the human occupant is determined to be seated in the vehicle seat, a height sensor to obtain height sensor data; and

statistically determine from the backrest sensor data, the base cushion sensor data, and the height sensor data, whether the human occupant matches a reference profile, the reference profile used as a baseline to identify the human occupant.

21. The at least one machine-readable medium of claim 20, wherein the instructions to determine that the human occupant is seated in the vehicle seat comprise instructions to:

obtain an image of the vehicle seat; and

use a facial detection process on the image to determine whether a face is present in the image.

22. The at least one machine-readable medium of claim 21, further comprising instructions to conditionally using the facial detection process when the base cushion sensor data indicates that more than a threshold weight is on the vehicle seat.

23. The at least one machine-readable medium of claim 20, wherein the reference profile comprises a reference backrest sensor data set, a reference base cushion sensor data set, and a reference height sensor data set, and wherein the instructions to statistically determine whether the human occupant matches the reference profile comprise instructions to:

obtain a backrest confidence value indicating the likelihood that the backrest sensor data corresponds to the reference backrest sensor data set;

obtain a base cushion confidence value indicating the likelihood that the base cushion sensor data corresponds to the reference base cushion sensor data set;

obtain a height confidence value indicating the likelihood that the seat height sensor data corresponds to the reference height sensor data set; and

aggregate the backrest confidence value, the base cushion confidence value, and the height confidence value to obtain a total confidence value.

24. The at least one machine-readable medium of claim 23, wherein the backrest sensor data comprises a set of sensor values corresponding to a plurality of sensors in the backrest sensor arrangement.

25. The at least one machine-readable medium of claim 24, wherein the instructions to obtain the backrest confidence value comprise instructions to:

calculate a sum of sums of dot products of the set of sensor values corresponding to the plurality of sensors in the backrest sensor arrangement and a set of sensor values corresponding to the reference backrest sensor data set; and

compare the sum of sums with a sum of sums of the set of sensor values corresponding to the reference backrest sensor data set with itself.