Add to Chrome
Add to Firefox
Add to EdgeTowards Dialogue-based Navigation with Multivariate Adaptation driven by Intention and Politeness for Social Robots
Sep 19, 2018
Service robots need to show appropriate social behavior in order to deploy in social environments such as healthcare, education, retail, etc. Some of the main capabilities that robots should have are navigation and conversational skill. If the person is impatient, he might want a robot to navigate faster and vice versa. Linguistic features that derive politeness can provide social cues about person's patient and impatient behavior. The novelty presented in this paper is to dynamically incorporate politeness in robotic dialogue systems for navigation. Understanding the politeness in users' speech can be used to modulate the robot behavior and responses. Therefore, we developed a dialogue system to navigate in an indoor environment, which produces different robot behaviors and responses based on users' intention and degree of politeness. We deploy and test our system with the Pepper robot that adapts to the changes in user's politeness.
