Personalised Human-Robot Co-Adaptation in Instructional Settings using Reinforcement Learning

In the domain of robotic tutors, personalised tutoring has started to receive scientists’ attention, but is still relatively underexplored. Previous work using reinforcement learning (RL) has addressed personalised tutoring from the perspective of affective policy learning. In this paper we build on previous work on affective policy learning that used RL to learn what robot’s supportive behaviours are preferred by users in an educational scenario. We propose a RL framework for personalisation that selects a robot’s supportive behaviours to maximize user’s task performance in a learning scenario where a Pepper robot acting as a tutor helps people learning how to solve grid-based logic puzzles. This work is relevant for the development of persuasive embodied agents and social robots used to support users in different scenarios. In particular, this paper makes a contribution towards the development of algorithms for human-robot co-adaptation that enable robots and agents to select effective strategies to establish long-term relationships with human users.