Air-conditioning control device and computer-readable recording medium

1 . An air-conditioning control device comprising a processor comprising hardware, the processor being configured to execute:

estimating an emotion of an occupant in a vehicle interior when a spot air conditioning is performed on the occupant;

controlling the spot air conditioning based on an estimation result using a machine learning model, the machine learning model is trained to receive a numerical value indicating the estimated emotion of the occupant as input, and to output an on-off timing of the spot air conditioning, a switching destination of a target of the spot air conditioning, a temperature of the spot air conditioning, or a wind direction of the spot air conditioning;

acquiring sensor data from a sensor installed in the vehicle interior; and

estimating at least one of a situation of the occupant, a situation of a passenger compartment, or a situation around the occupant from the acquired sensor data to acquire a situation estimation result of the at least one of the situation of the occupant, the situation of the passenger compartment, or the situation around the occupant;

wherein the input to the machine learning model further includes the situation estimation result.

2 . The air-conditioning control device according to claim 1 , wherein the estimating of the emotion of the occupant includes detecting a state of the occupant from the acquired sensor data, and estimating the emotion of the occupant from the acquired sensor data using the machine learning model.

3 . The air-conditioning control device according to claim 2 , wherein the sensor includes a camera installed in the vehicle interior, and the sensor data includes image data obtained by the camera.

4 . The air-conditioning control device according to claim 3 , wherein the estimating of the emotion of the occupant is based on an expression of the occupant included in the image data.

5 . The air-conditioning control device according to claim 2 , wherein the sensor includes a biometric sensor installed in the vehicle interior, and the sensor data includes biometric data obtained by the biometric sensor.

6 . The air-conditioning control device according to claim 5 , wherein the estimating of the emotion of the occupant is based on a body temperature, a heartbeat, a pulse, a blood pressure, or an electroencephalogram of the occupant included in the biometric data.

7 . The air-conditioning control device according to claim 2 , wherein the sensor includes a camera and a biometric sensor that are installed in the vehicle interior, and the sensor data includes image data obtained by the camera and biometric data obtained by the biometric sensor.

8 . The air-conditioning control device according to claim 7 , wherein the estimating of the emotion of the occupant is based on an expression of the occupant included in the image data and on a body temperature, a heartbeat, a pulse, a blood pressure, or an electroencephalogram of the occupant included in the biometric data.

9 . The air-conditioning control device according to claim 1 , wherein the estimating of the emotion of the occupant is repeatedly executed, and the controlling of the spot air-conditioning based on the estimation result includes adjusting a parameter of the spot air-conditioning such that a repeatedly obtained estimation result of the emotion of the occupant varies in a direction of comfort from discomfort.

10 . The air-conditioning control device according to claim 9 , wherein the parameter of the spot air-conditioning includes a target of the spot air conditioning, an air volume, a temperature, or a wind direction.

11 . The air-conditioning control device according to claim 1 , wherein the estimating of the at least one of the situation of the occupant, the situation of the passenger compartment, or the situation around the occupant is performed using the machine learning model to which the acquired sensor data is input as input data and from which the at least one of the situation of the occupant, the situation of the passenger compartment, or the situation around the occupant is output as output data.

12 . The air-conditioning control device according to claim 1 ,

wherein:

the sensor data includes image data of the occupant;

the processor is configured to execute estimating the situation of the occupant based on the image data of the occupant; and

the input to the machine learning model includes an estimation result of the situation of the occupant as the situation estimation result.

13 . The air-conditioning control device according to claim 12 , wherein the situation of the occupant is at least one of:

a degree of a facial shine of the occupant due to perspiration or moisture;

how a heard hair of the occupant is gathered due to perspiration or moisture;

whether or not a front hair of the occupant is caught in an eye of the occupant;

whether or not a body, clothing, or shoes of the occupant is wet with rain; and

material for a clothing of shoes of the occupant.

14 . The air-conditioning control device according to claim 1 , wherein:

the sensor data includes image data of the vehicle interior;

the processor is configured to execute estimating the situation of the passenger compartment based on the image data of the vehicle interior; and

the input to the machine learning model includes an estimation result of the situation of the passenger compartment as the situation estimation result.

15 . The air-conditioning control device according to claim 14 , wherein:

the situation of the passenger compartment is at least one of:

a presence or absence of a portion where direct sunlight is shining in the passenger compartment and its position; and

a presence or absence of a portion which is overheated in the passenger compartment and its position.

16 . The air-conditioning control device according to claim 1 , wherein:

the sensor data includes sensor data of surroundings of the vehicle;

the processor is configured to execute estimating the situation around the occupant based on the sensor data of the surroundings of the vehicle; and

the input to the machine learning model includes an estimation result of the situation around the occupant as the situation estimation result.

17 . The air-conditioning control device according to claim 16 , wherein the situation around the occupant is at least one of:

whether or not surrounding atmosphere of the occupant is dry; and

an ambient air temperature of the occupant.

18 . A non-transitory computer-readable recording medium with an executable program stored thereon, the program causing a processor to execute:

estimating an emotion of an occupant in a vehicle interior when a spot air conditioning is performed on the occupant;

controlling the spot air conditioning based on an estimation result using a machine learning model, the machine learning model is trained to receive a numerical value indicating the estimated emotion of the occupant as input, and to output an on-off timing of the spot air conditioning, a switching destination of a target of the spot air conditioning, a temperature of the spot air conditioning, or a wind direction of the spot air conditioning;

acquiring sensor data from a sensor installed in the vehicle interior; and

estimating at least one of a situation of the occupant, a situation of a passenger compartment, or a situation around the occupant from the acquired sensor data to acquire a situation estimation result of the at least one of the situation of the occupant, the situation of the passenger compartment, or the situation around the occupant;

wherein the input to the machine learning model further includes the situation estimation result.