IN-VEHICLE THERMAL DISTRESS MITIGATION

1 . A method for in-vehicle thermal distress mitigation, comprising:

monitoring, by one or more processors using one or more temperature sensors, an in-vehicle temperature of a vehicle;

receiving, by the one or more processors from a camera mounted to or in the vehicle, one or more images of an environment within the vehicle depicting a driver of the vehicle;

responsive to determining the in-vehicle temperature of the vehicle satisfies a temperature threshold, executing, by the one or more processors, a machine learning model using at least the one or more images to detect the driver is in a thermal distress state;

responsive to determining the driver is in the thermal distress state, generating, by the one or more processors, a message in the vehicle requesting driver input; and

when no affirmative response to the message is received within a defined time period subsequent to the generation of the message, generating, by the one or more processors, a real-time video feed depicting the driver to a remote computing system.

2 . The method of claim 1 , further comprising:

determining, by the one or more processors, the vehicle is occupied,

wherein executing the machine learning model using the one or more images is further responsive to determining the vehicle is occupied.

3 . The method of claim 2 , wherein determining the vehicle is occupied comprises determining, by the one or more processors, the vehicle is occupied using one or more occupancy sensors.

4 . The method of claim 1 , further comprising:

determining, by the one or more processors, the vehicle has been stationary for at least a defined duration,

wherein executing the machine learning model using the one or more images is further responsive to determining the vehicle has been stationary for at least the defined duration.

5 . The method of claim 4 , further comprising:

determining, by the one or more processors, the vehicle is occupied,

wherein executing the machine learning model using the one or more images is further responsive to determining the vehicle is occupied and determining the vehicle has been stationary for at least the defined duration.

6 . The method of claim 1 , wherein executing the machine learning model using at least the one or more images to detect the driver is in the thermal distress state comprises:

generating, by the one or more processors using the machine learning model, a confidence score indicating a likelihood that the driver is in the thermal distress state; and

determining, by the one or more processors, the confidence score satisfies a threshold.

7 . The method of claim 1 , further comprising:

detecting, by the one or more processors, one or more driver actions over a time period prior to determining the in-vehicle temperature exceeds the temperature threshold,

wherein executing the machine learning model using the one or more images comprises executing, by the one or more processors, the machine learning model using the one or more images and identifications of the one or more driver actions over the time period to detect the driver is in the thermal distress state.

8 . The method of claim 7 , wherein detecting the one or more driver actions over the time period comprises detecting the driver taking a drink of a liquid and the driver exiting and reentering the vehicle,

wherein executing the machine learning model using the one or more images comprises executing, by the one or more processors, the machine learning model using the one or more images and identifications of a number of instances in which the driver exited and reentered the vehicle within the time period and a number of instances in which the driver drank a liquid within the time period to detect the driver is in the thermal distress state.

9 . The method of claim 1 , further comprising:

detecting, by the one or more processors, a change in state of the vehicle from a mobile state to a stationary state,

wherein executing the machine learning model using the one or more images comprises executing, by the one or more processors, the machine learning model using the one or more images and an identification of the change in state of the vehicle from the mobile state to the stationary state to detect the driver is in the thermal distress state.

10 . The method of claim 1 , further comprising:

detecting, by the one or more processors, an outside air temperature,

wherein executing the machine learning model using the one or more images comprises executing, by the one or more processors, the machine learning model using the one or more images and the outside air temperature to detect the driver is in the thermal distress state.

11 . The method of claim 1 , further comprising:

responsive to determining the in-vehicle temperature of the vehicle satisfies the temperature threshold, detecting, by the one or more processors, a state of one or more devices within the vehicle,

wherein executing the machine learning model using the one or more images comprises executing, by the one or more processors, the machine learning model using the one or more images and the state of the one or more devices within the vehicle.

12 . The method of claim 11 , wherein detecting the state of the one or more devices within the vehicle comprises detecting an opening state of one or more windows within the vehicle or an activation state of an air conditioning system within the vehicle,

wherein executing the machine learning model using the one or more images comprises executing, by the one or more processors, the machine learning model using the one or more images and the opening state of the one or more windows and the activation state of the air conditioning system.

13 . The method of claim 1 , wherein generating the message comprises:

selecting, by the one or more processors, a pre-recorded audio message from memory; and

playing, by the one or more processors, the pre-recorded audio message through a speaker of the vehicle.

14 . The method of claim 1 , wherein receiving the one or more images comprises receiving, by the one or more processors, a sequence of images depicting the driver over time,

wherein executing the machine learning model using the one or more images comprises executing, by the one or more processors, the machine learning model using the one or more images based on a change in pose of the driver over the sequence of images.

15 . The method of claim 1 , wherein receiving the one or more images comprises receiving, by the one or more processors, at least one image depicting a facial expression of the driver,

wherein executing the machine learning model using the one or more images comprises executing, by the one or more processors, the machine learning model using the one or more images based on the depicted facial expression of the driver.

16 . The method of claim 1 , wherein executing the machine learning model using the one or more images comprises executing, by the one or more processors, the machine learning model using the one or more images generated prior to the generation of the message.

17 . The method of claim 1 , further comprising:

receiving, by the one or more processors from the camera mounted to or in the vehicle, second one or more images of the environment within the vehicle depicting the driver of the vehicle;

responsive to determining a second in-vehicle temperature of the vehicle satisfies the temperature threshold, executing, by the one or more processors, the machine learning model using at least the one or more second images to detect the driver is in the thermal distress state;

responsive to determining the driver is in the thermal distress state, generating, by the one or more processors, a second message in the vehicle requesting driver input;

detecting, by the one or more processors, the requested driver input; and

pausing, by the one or more processors, monitoring of the in-vehicle temperature for a second defined time period before initiating monitoring the in-vehicle temperature again.

18 . A system for in-vehicle thermal distress mitigation, comprising:

one or more processors configured by instructions stored in memory to:

monitor, using one or more temperature sensors, an in-vehicle temperature of a vehicle;

receive, from a camera mounted to or in the vehicle, one or more images of an environment within the vehicle depicting a driver of the vehicle;

responsive to determining the in-vehicle temperature of the vehicle satisfies a temperature threshold, execute a machine learning model using at least the one or more images to detect the driver is in a thermal distress state;

responsive to determining the driver is in the thermal distress state, generate a message in the vehicle requesting driver input; and

when no affirmative response to the message is received within a defined time period subsequent to the generation of the message, generate a real-time video feed depicting the driver to a remote computing system.

19 . The system of claim 18 , wherein the one or more processors are further configured to:

determine the vehicle is occupied,

wherein the one or more processors are configured to execute the machine learning model using the one or more images further responsive to determining the vehicle is occupied.

20 . The system of claim 19 , wherein the one or more processors are configured to determine the vehicle is occupied by determining the vehicle is occupied using one or more occupancy sensors.