Information processing apparatus, mobile object, control method thereof, and storage medium for causing mobile device to move to meeting location with user

A method of controlling an information processing apparatus, the method comprising:

• acquiring, from a communication device of a user, first utterance information by the user;
• specifying a predetermined region according to a mark included in the first utterance information;
• acquiring a captured image captured around the specified predetermined region;
• acquiring a movement direction of the user from the first utterance information, and set a probability distribution indicating probabilities that the user exists to each of a plurality of regions, wherein the plurality of regions are areas that have been divided from the predetermined region;
• acquiring, from a communication device of the user, second utterance information by the user;
• analyzing the captured image and extract one or more candidate users and each feature of the one or more candidate users;
• generating a question inquiring a feature of the user based on the extracted one or more candidate users and each feature, and send the generated question,
• acquiring, from a communication device of the user, second utterance information for the question and by the user;
• presuming the user based on the set probability distribution, the extracted feature of the one or more candidate users and each feature, and the second utterance information, and
• sending, to a moving object, a meeting position determined based on the presumed user to cause the moving object to move the meeting position.

A mobile object comprising:

• a communication device that communicates with a communication device of a user;
• a camera that images a surrounding portion of the mobile object;
• a storage device that stores instructions;
• and
• at least one processor that executes the instructions to:
• acquire, from the communication device of the user and by the communication device, first utterance information by the user;
• specify a predetermined region according to a mark included in the first utterance information;
• acquire a captured image captured around the specified predetermined region;
• acquire a movement direction of the user from the first utterance information, and set a probability distribution indicating probabilities that the user exists to each of a plurality of regions, wherein the plurality of regions are areas that have been divided from the predetermined region;
• analyze the captured image and extract one or more candidate users and each feature of the one or more candidate users;
• and
• generate a question inquiring a feature of the user based on the extracted one or more candidate users and each feature, and send the generated question,
• acquire, from a communication device of the user and by the communication device, second utterance information for the question and by the user;
• presume the user based on the set probability distribution, the extracted feature of one or more persons, and the second utterance information, and
• control the moving object to move to a meeting position determined based on the presumed user.

A method of controlling a mobile object including a communication device that communicates with a communication device of a user and a camera that images a surrounding portion of the mobile object, the method comprising:

• acquiring, from the communication device of the user and by the communication device, first utterance information by the user;
• specifying a predetermined region according to a mark included in the first utterance information;
• acquiring a captured image captured around the specified predetermined region;
• acquiring a movement direction of the user from the first utterance information, and set a probability distribution indicating probabilities that the user exists to each of a plurality of regions, wherein the plurality of regions are areas that have been divided from the predetermined region;
• analyzing the captured image and extract a feature of one or more candidate users and each feature of the one or more candidate users;
• generating a question inquiring a feature of the user based on the extracted one or more candidate users and each feature, and send the generated question;
• acquiring, from a communication device of the user and by the communication device, second utterance information for the question and by the user;
• presuming the user based on the set probability distribution, the extracted feature of one or more persons, and the second utterance information, and
• controlling the moving object to move to a meeting position determined based on the presumed user.