A Context-Aware Adaptability Model for Service Robots

This article presents a context-aware, adaptable service selection model for a social robot, giving it the ability to estimate the user’s expectation, assess the degree of satisfaction and use it as feedback to improve subsequent interactions. We established specific measures for expectation and satisfaction, estimated using Bayesian inference and used to control the human-robot interaction. This work is proposed to overcome the fact that service robots are usually designed to perform within a very strict operational envelope, sometimes requiring all knowledge to be known and locally preprogrammed. The model was tested in demanding simulated scenarios, showing promising results, and also in exploratory experiments with users.