Ankuj Arora - Action Model Learning for Socio-Communicative Human-Robot Interaction

Driven with the objective of rendering robots as socio-communicative, there has been a heightened interest towards researching techniques to endow robots with social skills and “commonsense” in order to render them acceptable. This “commonsense” is, however, not so common, as even a standard dialogue exchanges integrates behavioral subtleties that are difficult to codify. In such a scenario, learning the behavioral model of the robot is a promising approach. This study tries to solve the problem of learning robot behavioral models in the Automated Planning and Scheduling (APS) paradigm of Artificial Intelligence. During the course of this study, we principally introduce two symbolic and deep learning systems, by the names SPMSAT and PDeepLearn respectively, which facilitate the learning of action models, and extend the scope of these new techniques to learn robot behavioral models in a Human-Robot Interaction (HRI) scenario. The long term objective is to empower robots to communicate autonomously with humans sans the need of “wizard” intervention.