Multiple Script-based Task Model and Decision/Interaction Model for Fetch-and-carry Robot

Abstract

The characteristics of the behaviors of home- service robots are different from those of other intelligent systems. Based on these characteristics, we developed a task model that can both support meaningful interactions and improve task performance. The task model is able to generate the robot's diverse actions and aid its interaction management. A decision/interaction model was established to decide autonomy of decision problems that are embedded on the task model. When Interactions with humans are required, it also manages timing of the interactions. We demonstrated the applicability of the developed system with a computer simulation.