Modeling Basic Aspects of Cyber-Physical Systems, Part II

TL;DR

This paper extends the modeling framework for cyber-physical systems by analyzing complex case studies, demonstrating the core language's scalability and the need for static partial derivatives support for better modeling of rigid body dynamics.

Contribution

It shows how a small core language can model complex CPS like quadcopters and robots, and highlights the importance of adding static partial derivatives for rigid body dynamics.

Findings

Core language supports larger CPS models than previously shown.

Static partial derivatives are essential for effective rigid body dynamics modeling.

Enhanced language features improve expressiveness for complex systems.

Abstract

We continue to consider the question of what language features are needed to effectively model cyber-physical systems (CPS). In previous work, we proposed using a core language as a way to study this question, and showed how several basic aspects of CPS can be modeled clearly in a language with a small set of constructs. This paper reports on the result of our analysis of two, more complex, case studies from the domain of rigid body dynamics. The first one, a quadcopter, illustrates that previously proposed core language can support larger, more interesting systems than previously shown. The second one, a serial robot, provides a concrete example of why we should add language support for static partial derivatives, namely that it would significantly improve the way models of rigid body dynamics can be expressed.