Position detecting device, autonomous mobile device, method, and computer program product

An autonomous mobile device comprising:

• a storage unit that stores therein:
• identification information assigned to each sound source that can be visually and auditorily identified,
• image processing information including an image processing procedure for visually identifying sound sources and visual feature information indicative of visual features of the sound sources,
• sound processing information including a sound processing procedure for auditorily identifying the sound sources and auditory feature information indicative of auditory features of the sound sources, and
• a plurality of pieces of detection strategy information indicative of which one or both of the image processing information and the sound processing information is to be used to detect the sound sources;
• an image acquiring unit that acquires an image of the sound sources;
• a sound collecting unit that collects sound produced by the sound sources;
• a visual-feature detecting unit that detects a visual feature of a sound source by performing the image processing procedure based on the image acquired by the image acquiring unit and the image processing information, detects a first position of the sound source from the visual feature, and outputs visually localizing information indicative of the first position of the sound source;
• an auditory-feature detecting unit that detects an auditory feature of the sound source by performing the sound processing procedure based on the sound collected by the sound collecting unit and the sound processing information, detects a second position of the sound source based on the auditory feature, and outputs auditorily localizing information indicative of the second position of the sound source;
• a sound-source detecting unit that controls the visual-feature detecting unit and the auditory-feature detecting unit based on the detection strategy information, and detects a third position of the sound source based on the visually localizing information output from the visual-feature detecting unit and the auditorily localizing information output from the auditory-feature detecting unit;
• and
• a positional-relation controlling unit that controls positional relation of the device with the sound source based on the position of the sound source detected by the sound-source detecting unit.

A method of detecting a position of a sound source that can be visually and auditorily identified, comprising:

• acquiring an image of sound sources;
• collecting sound produced by the sound sources;
• first detecting including detecting a visual feature of a sound source based on:
• identification information assigned to each sound source that can be visually and auditorily identified,
• image processing information including an image processing procedure for visually identifying the sound sources and visual feature information indicative of visual features of the sound source,
• sound processing information including a sound processing procedure for auditorily identifying the sound sources and auditory feature information indicative of auditory features of the sound source,
• a plurality of pieces of detection strategy information indicative of which one or both of the image processing information and the sound processing information is to be used to detect the sound source, and
• the acquired image;
• second detecting including detecting a first position of the sound source from the visual feature detected at the first detecting;
• first outputting including outputting visually localizing information indicative of the first position of the sound source detected at the second detecting;
• third detecting including detecting an auditory feature of the sound source based on the sound processing information and the collected sound;
• fourth detecting including detecting a second position of the sound source based on the auditory feature detected at the third detecting;
• second outputting including outputting auditorily localizing information indicative of the second position of the sound source detected at the fourth detecting;
• and
• fifth detecting including detecting a third position of the sound source based on the visually localizing information output at the first outputting and the auditorily localizing information output at the second outputting.

A computer program product having a computer readable medium including program instructions for detecting a position of a sound source that can be visually and auditorily identified, wherein the instructions, when executed by a computer, cause the computer to perform:

• acquiring an image of sound sources;
• collecting sound produced by the sound sources;
• first detecting including detecting a visual feature of a sound source based on:
• identification information assigned to each sound source that can be visually and auditorily identified,
• image processing information including an image processing procedure for visually identifying the sound sources and visual feature information indicative of visual features of the sound source,
• sound processing information including a sound processing procedure for auditorily identifying the sound sources and auditory feature information indicative of auditory features of the sound source,
• a plurality of pieces of detection strategy information indicative of which one or both of the image processing information and the sound processing information is to be used to detect the sound source, and
• the acquired image;
• second detecting including detecting a first position of the sound source from the visual feature detected at the first detecting;
• first outputting including outputting visually localizing information indicative of the first position of the sound source detected at the second detecting;
• third detecting including detecting an auditory feature of the sound source based on the sound processing information and the collected sound;
• fourth detecting including detecting a second position of the sound source based on the auditory feature detected at the third detecting;
• second outputting including outputting auditorily localizing information indicative of the second position of the sound source detected at the fourth detecting;
• and
• fifth detecting including detecting a third position of the sound source based on the visually localizing information output at the first outputting and the auditorily localizing information output at the second outputting.