Incrementality, Grounding and Fluidity for Human-Robot Interaction

Developing suitable communicative grounding mechanisms for communication in the sense of (Clark and Brennan, 1991) is an ongoing challenge for designers of robotic systems which interpret speech. If grounding is the way in which interaction participants build and align their internal representations towards shared information or 'common ground', given the vastly different internal representations of humans and robots, one might concede the title of Kruijff (2012)'s paper: 'There is no common ground in human-robot interaction'.

However despite the lack of 'real' common ground, a robot can still understand what the user means 'to a criterion sufficient for current purposes' (Clark and Brennan, 1991) at a given point in the interaction, if it is equipped with grounding mechanisms which deal with the inherent uncertainty in situated dialogue for a robot.

I will present a simple real-time, real-world grounding framework we are developing in Bielefeld's Dialogue System's Group, and a system which implements it in a simple robot, allowing investigation into different grounding strategies. We put particular focus on the grounding effects of non-linguistic task-related actions, and how these are achieved in real time. We experiment with a trade-off between the fluidity of the grounding mechanism with the 'safety' of ensuring task success. The framework consists of a combination of interactive Harel statecharts and the Incremental Unit framework (Schlangen and Skantze, 2011). We evaluate its in-robot implementation in a study with human users and find that in simple grounding situations, a model allowing greater fluidity is perceived to have better understanding of the user's speech.