A General Framework of Task Understanding for Tour-Guide Robots in Exhibition Environments

Tour-guide robots are a kind of service robots that attract more and more attentions in hospitality industry, such as the museums and exhibitions in domestic or open environments. Two of their essential capabilities are to perform physical movements and to interact with visitors verbally and non-verbally. In this paper, we propose a general framework of the task understanding sub-system inspired by the Tsvetkov's cognitive information model, to formalize the cognitive information processing, and to enhance the cognitive architecture of tour-guide robots. The proposed framework follows the modular design principle, and consists of different modules that relate to the "thinking" process. We also provide a demonstrative implementation of the proposed framework with a tour-guide robot prototype, which is deployed in a real-world exhibition hall. The experimental results show that the proposed framework enables the tour guide robot to understand the stimuli from visitors.