A Cybernetic Approach to Characterization of Complex Sensory Environments: Implications for Human Robot Interaction

Abstract

Humans are increasingly interacting and collaborating with robotic and intelligent agents. How to make these interactions as effective as possible remains, however, an open question. Here, we argue that consistent understandings of the environment on the part of the human and agent are critical for their interaction and basing these understandings on only the objective features of sensory inputs may be inadequate. To that end, the current paper presents a novel approach to more integrated characterizations of the sensory environment that encompass objective and subjective features of sensory inputs. We propose that an approach to signal and behavioral estimation consistent with the control and communication theoretic perspective of Cybernetics could inform human robot interaction (HRI) applications. Specifically, we offer a potential path forward for quantifying similarity in stimulus events that can lead to consistent understandings of the environment, which when applied to HRI can enhance human-agent communication in HRI applications.