Adjustment device, adjustment system, and adjustment method

An adjustment device that adjusts an estimation method of a state of an occupant in at least one occupant state estimating device that includes a plurality of occupant state estimating devices and estimates the state of the occupant, the adjustment device comprising:

• processing circuitry configured to collect estimation results of the state of the occupant estimated by the plurality of occupant state estimating devices;
• determine whether or not it is necessary to check the state of the occupant on a basis of the collected estimation results and a machine learning model that outputs checking necessity information indicating whether or not it is necessary to check the state of the occupant using the estimation results as an input;
• determine a behavior of the occupant on a basis of behavior determination information for determining the behavior of the occupant and set the state of the occupant when the processing circuitry determines that it is necessary to check the state of the occupant;
• output adjustment information for causing the at least one occupant state estimating device to adjust a state estimation condition used for estimating the state of the occupant in such a manner that the state of the occupant is estimated as the state of the occupant having been set;
• collect, as reference information, at least one of vehicle interior environment determination information for determining an environment in a vehicle and abnormal-time motion determination information for determining a motion that the occupant tends to make in a case where the occupant is in an abnormal state;
• and
• determine whether or not it is necessary to check the state of the occupant on a basis of the collected estimation results, the collected reference information, and the machine learning model that uses the estimation results and the reference information as inputs and outputs the checking necessity information.

An adjustment method that adjusts an estimation method of a state of an occupant in at least one occupant state estimating device that includes a plurality of occupant state estimating devices and estimates the state of the occupant, the adjustment method comprising:

• collecting estimation results of the state of the occupant estimated by the plurality of occupant state estimating devices;
• determining whether or not it is necessary to check the state of the occupant on a basis of the collected estimation results and a machine learning model that outputs checking necessity information indicating whether or not it is necessary to check the state of the occupant using the estimation results as an input;
• determining a behavior of the occupant on a basis of behavior determination information for determining the behavior of the occupant and setting the state of the occupant when it is determined that it is necessary to check the state of the occupant;
• outputting adjustment information for causing the at least one occupant state estimating device to adjust a state estimation condition used for estimating the state of the occupant in such a manner that the state of the occupant is estimated as the state of the occupant having been set;
• collecting, as reference information, at least one of vehicle interior environment determination information for determining an environment in a vehicle and abnormal-time motion determination information for determining a motion that the occupant tends to make in a case where the occupant is in an abnormal state;
• and
• determining whether or not it is necessary to check the state of the occupant on a basis of the collected estimation results, the collected reference information, and the machine learning model that uses the estimation results and the reference information as inputs and outputs the checking necessity information.