Social Robots Privacy Enhancement Using Colored Petri Net (CPN) for Behavior Modeling: A Case Study of Asus Zenbo Robot

Abstract

The interactions between a social robot and the user consist of continuous communication and behavior involving different data types that can be subject to cybersecurity attacks and prone to user privacy concerns. In this paper, we use Colored Petri Net (CPN) to develop two graphical models that illustrate different patterned behavior of a robot during such interaction. Using CPN, we model and analyze complicated robot system interactions considering synchronization and concurrency of events that can be subject to privacy violations. We focus on two specific scenarios involving user registration and medication reminder to provide an efficient illustration of the objects and events collaborated to carry out the intended tasks of the robot pertaining to privacy issues. The results show that CPN modeling simulation can capture and represent the robot's behavior, provide a better understanding of the task execution, and highlight users' privacy gaps requiring immediate controls.