Integrating vision and audition within a cognitive architecture to track conversations

Author

Trafton, J. Gregory ; Bugajska, Magdalena D. ; Fransen, Benjamin R. ; Ratwani, Raj M.

Author_Institution

Naval Res. Lab., Washington, DC, USA

fYear

2008

fDate

12-15 March 2008

Firstpage

201

Lastpage

208

Abstract

We describe a computational cognitive architecture for robots which we call ACT-R/E (ACT-R/Embodied). ACT-R/E is based on ACT-R [1, 2] but uses different visual, auditory, and movement modules. We describe a model that uses ACT-R/E to integrate visual and auditory information to perform conversation tracking in a dynamic environment. We also performed an empirical evaluation study which shows that people see our conversational tracking system as extremely natural.

Keywords

audio signal processing; cognitive systems; human-robot interaction; robot vision; tracking; ACT-R/Embodied; auditory information; auditory modules; computational cognitive architecture; conversation tracking; conversational tracking system; dynamic environment; empirical evaluation study; integrating audition; integrating vision; movement modules; robots; visual information; visual modules; Cameras; Facial animation; Humans; Robot vision systems; Visualization; ACT-R; Cognitive modeling; Conversation following; human-robot interaction;

fLanguage

English

Publisher

ieee

Conference_Titel

Human-Robot Interaction (HRI), 2008 3rd ACM/IEEE International Conference on

Conference_Location

Amsterdam

ISSN

2167-2121

Print_ISBN

978-1-60558-017-3

Type

conf

Filename

6249436