Machine learning for olfactory mood alteration

• US 10,150,351 B2
• Filed: 02/08/2017
• Issued: 12/11/2018
• Est. Priority Date: 02/08/2017
• Status: Active Grant

• Current Assignee: LP-Research Inc.
• Original Assignee: LP-Research Inc.
• Inventors: Lin, Zhuohua, Petersen, Klaus, Schluter, Tobias, Shiurba, Robert, Yap, Huei Ee
• Primary Examiner(s): Chen, Xuemei G
• Application Number:
  US15/427,596
• Publication Number:
  US 20180225551A1
• Time in Patent Office: 671 Days
• Field of Search: 180272, 340441, 340576, 348 77, 352 85, 382104, 382118, 463 16, 700 94, 701 1, 701 36, 701 48, 705 1419, 705 1442, 713186
• CPC Class Codes: A61L 2209/111 Sensor means, e.g. motion, brightness, scent, contaminant sensors
• CPC Class Codes: A61L 9/122 comprising a fan
• CPC Class Codes: A61L 9/14 using sprayed or atomised substances including air-liquid contact processes
• CPC Class Codes: B60H 1/00357 Air-conditioning arrangements specially adapted for particular vehicles
• CPC Class Codes: B60H 3/0035 characterised by the control methods for adding the substance
• CPC Class Codes: G06V 20/597 Recognising the driver's state or behaviour, e.g. attention or drowsiness

Abstract

System, method and media for altering the mood of an occupant (such as a driver) of a vehicle. Reckless operation of motor vehicles by emotionally disturbed drivers is a major cause of traffic accidents just like alcohol, drug, and cell phone use. Emotional states such as annoyance, anger, anxiety, depression, and feeling hurried can significantly impair awareness by slowing observation and reaction times. Scents, both pleasant and unpleasant, have major effects on mood and sense of well-being. Accordingly, embodiments of the invention provide for an adaptive system which can detect a driver'"'"'s mood, disperse an appropriate scent to improve the mood if it is unsafe, and learn the impact of scents on different users to effectively improve their moods.

Claims

1. A system for altering a mood of an occupant of a vehicle, comprising:

one or more cameras located in the vehicle and configured to capture facial imagery of the occupant of the vehicle;

one or more scent dispersion units, each scent dispersion unit configured to disperse a scent perceptible to the occupant of the vehicle;

a processor;

one or more non-transitory computer storage media storing;

one or more occupant profiles, wherein each occupant profile comprises mood classification information and mood actuation information; and

computer-executable instructions that, when executed by the processor, perform a method of altering the mood of the occupant of the vehicle, the method comprising the steps of;

determining an identity of the occupant of the vehicle;

loading an occupant profile for the occupant of the vehicle from the one or more occupant profiles;

obtaining initial facial imagery of the occupant from the one or more cameras;

identifying, based at least in part on the initial facial imagery of the occupant and the mood classification information, a current mood for the occupant;

determining, from the mood actuation information, a mood actuation to change the current mood to a desired mood;

dispersing one or more scents associated with the mood actuation;

obtaining updated facial imagery of the occupant from the one or more cameras;

identifying, based at least in part on the updated facial imagery of the occupant and the mood classification information, an updated mood for the occupant; and

updating the mood actuation information based on the updated mood for the occupant and the dispersed one or more scents through a machine learning model.

2. The system of claim 1, wherein the identity of the occupant is determined by recognizing a face of the occupant in the facial imagery of the occupant.

3. The system of claim 1, wherein the method further comprises the step of updating the mood classification information based on the updated facial imagery and the mood actuation.

4. The system of claim 1, wherein the system further comprises one or more supplementary sensors, and wherein the current mood is determined in part based on data from the supplementary sensors.

5. The system of claim 4, wherein the one or more supplementary sensors include a heart-rate sensor for the occupant.

6. The system of claim 1, wherein the system further includes one or more onboard automotive sensors, and wherein current mood is determined in part based on data from the one or more onboard automotive sensors.

7. The system of claim 1, wherein the current mood indicates that the occupant is angry and the desired mood is a calm mood.

8. The system of claim 1, wherein the current mood indicated that the occupant is distracted and the desired mood is an alert mood.

9. A method of altering a mood of an occupant of a vehicle, comprising the steps of:

determining an identity of the occupant of the vehicle;

obtaining initial facial imagery of the occupant from one or more cameras in the vehicle;

identifying, based at least in part on the initial facial imagery of the occupant and mood classification information for the occupant, a current mood for the occupant;

determining, from mood actuation information for the occupant, a mood actuation to change the current mood to a desired mood;

dispersing one or more scents associated with the mood actuation;

obtaining updated facial imagery of the occupant from the one or more cameras;

identifying, based at least in part on the updated facial imagery of the occupant and the mood classification information, an updated mood for the occupant; and

updating the mood classification information based on the updated facial imagery and the mood actuation through a machine learning model.

10. The method of claim 9, further comprising the step of updating the mood actuation information based on the updated mood for the occupant and the dispersed one or more scents.

11. The method of claim 9, wherein the identity of the occupant is determined based on a transponder carried by the occupant.

12. The method of claim 9, wherein the current mood is identified based in part on physiological data for the occupant obtained from one or more supplementary sensors in the vehicle.

13. The method of claim 12, wherein the physiological data is heart rate data and wherein the one or more supplementary sensors include a heart-rate sensor.

14. The method of claim 9, wherein the current mood is identified based in part on driving data for the vehicle obtained from one or more onboard vehicular sensors.

15. The method of claim 9, wherein the current mood is excited, and the desired mood is calm.

16. One or more computer storage media storing computer-executable instructions that, when executed by a processor, perform a method of altering a mood of an occupant of a vehicle, the method comprising the steps of:

determining an identity of the occupant of the vehicle;

obtaining initial facial imagery of the occupant from one or more cameras in the vehicle;

obtaining physiological data for the occupant from one or more physiological sensors in the vehicle;

identifying, based at least in part on the initial facial imagery of the occupant, the physiological data for the occupant, and mood classification information for the occupant, a current mood for the occupant;

determining, from mood actuation information for the occupant, a mood actuation to change the current mood to a desired mood;

dispersing one or more scents associated with the mood actuation;

obtaining updated facial imagery of the occupant from the one or more cameras;

identifying, based at least in part on the updated facial imagery of the occupant and the mood classification information, an updated mood for the occupant;

updating the mood actuation information based on the updated mood for the occupant and the dispersed one or more scents through a first machine learning model; and

updating the mood classification information based on the updated facial imagery and the mood actuation through a second machine learning model.

17. The media of claim 16, wherein the identity of the occupant is determined using facial recognition on the initial facial imagery of the occupant.

18. The media of claim 16, wherein the method further comprises the step of obtaining driving data from one or more onboard vehicular sensors, and wherein the current mood is determined in part based on the driving data.

19. The media of claim 16, wherein the media further stores a plurality of personalized occupant profiles, each personalized occupant profile comprising mood classification data and mood actuation data for a different occupant.

20. The media of claim 16, wherein the one or more scents dispersed are further associated with an additional mood actuation profile of an additional personalized occupant profile for an additional occupant of the vehicle.