Development of a Tour–Guide Robot Using Dialogue Models and a Cognitive Architecture

Integration of the Multi-DOA Estimation Functionality to Human-Robot Interaction

“…This interpretation is achieved by first obtaining an internal codification of the input [22] (i.e., the recognition phase), and then providing a meaning to that representation (i.e., the interpretation phase), guided by the contents of its memory and the current expectations of the DM. With this architecture, complex objectives can be fulfilled, such as a tour-guide robot capable of guiding a poster session [1].…”

“…There are several tables in the restaurant among which several users are seated. The location of the tables are known a priori to the robot 1 . When the robot is not doing anything, it is positioned in a place that is visually accessible to the users near the tables.…”

Robotic Orientation towards Speaker for Human-Robot Interaction

“…the Interpretation phase), guided by the contents of the episodic memory of the architecture and the current expectations of the DM. With this architecture, complex objectives can be fulfilled, such as a tour-guide robot capable to guide a poster session [1] and a spanish-spoken "guess the card" game [8].…”

“…This characteristic can be used by the Interpretation phase to enrich the meaning of the information received by other modalities, such as in the example presented in [1], where the tour-guide robot could approach and greet the user at the start of the tour, triggered by sound alone (instead of manual means); the robot could also face the user when being talked to during the tour, enhancing the 'naturality' of the conversation. The "direction" characteristic may also be used directly by the DM, as in the case of the Marco Polo example, where such information is essential for carrying out HRI.…”

“…The feature most used for DOA estimation when using a 2-mic array is the ITD, as there is a direct relation between the two, described in equation (1).…”