Cognitive Cyber-Physical System (C-CPS) for Human-Robot Collaborative Manufacturing

Abstract

In this paper, the fundamental concept of the newly coined framework C-CPS (cognitive cyber-physical system) is introduced with application to human-robot collaborative manufacturing. A human-robot hybrid cell is developed where a human and a robot collaborate to perform assembly of components in manufacturing. Human and robot's mutual trust, an expression of human and machine cognition, is modeled and computed in realtime for the collaborative task. In the proposed C-CPS framework, a cyber system (software system) along with necessary hardware (sensors) computes the mutual trust in real-time based on the performance and precision of the physical system (human and robot). The computed mutual trust is then communicated to the robot via the software system and to the human via a visual interface. The computed trust values are also used to control the motion of the robot based on an algorithm, and to stimulate the action of the human based on warnings communicated to the human via the visual interface. The performance of the physical system also impacts the cyber system (computing), and the vice versa. The framework is implemented with an actual human-robot collaborative assembly task. The experimental evaluation results show satisfactory performance of the proposed C-CPS in term of human-system interaction, manufacturing efficiency and quality.