Data driven discovery of cyber physical systems.

Area of Science:

• Computer Science
• Engineering
• Systems Science

Background:

• Cyber-physical systems (CPS) integrate computational elements with physical processes, finding applications in smart grids, robotics, and manufacturing.
• The inherent complexity of CPS, stemming from the interplay between cyber and physical components, poses significant modeling challenges.

Purpose of the Study:

• To propose a general, data-driven framework for discovering and modeling cyber-physical systems.
• To enable understanding of underlying mechanisms and prediction of system behavior.

Main Methods:

• A novel framework is presented for direct data-driven discovery of cyber-physical systems.
• The methodology includes identification of physical system components and inference of their transition logic.

Main Results:

• The proposed framework has been successfully applied to various real-world cyber-physical system examples.
• Demonstrated ability to model complex system dynamics from data.

Conclusions:

• The framework provides essential insights for assessing cyber-physical system performance.
• Facilitates debugging during implementation and guides system redesign for optimal performance.