method of conducting multimodal dialogue between a humanoid robot and a user, computer program and humanoid robot to implement such a method

摘要

a method of conducting a dialogue between a humanoid robot (r) and at least one user (u), comprising the following steps, performed iteratively by such a humanoid robot: i) acquisition of various input signals (s1, s2) from the respective sensors (c1, c2), at least one such sensor being a sound sensor and at least one other sensor being a motion or image sensor; ii) interpretation of acquired signals to recognize various events (evi) generated by such user, selected from a group comprising: the pronunciation of at least one word or sentence, one voice intonation, one gesture, one body posture, one facial expression; iii) determining a response from such a humanoid robot, comprising at least one event (evo) selected from a group comprising: the pronunciation of at least one word or sentence, a voice intonation, a gesture, a body posture, a facial expression ; iv) generation by such humanoid robot of such or each such event; characterized in that such step iii) comprises the determination of such a response as a function of at least two events generated jointly by such user and recognized in such step ii), of which at least one is not a word or sentence pronounced by such user . a computer program and a humanoid robot for performing and such a method.