A categorical approach to open and interconnected dynamical systems
In his 1986 Automatica paper Willems introduced the influential behavioural approach to control theory with an investigation of linear time-invariant (LTI) discrete dynamical systems. The behavioural approach places open systems at its centre, modelling by tearing, zooming, and linking. We show that...
Saved in:
| Published in: | Proceedings of the 31st Annual ACM/IEEE Symposium on Logic in Computer Science pp. 495 - 504 |
|---|---|
| Main Authors: | , , |
| Format: | Conference Proceeding |
| Language: | English |
| Published: |
New York, NY, USA
ACM
05.07.2016
|
| Series: | ACM Conferences |
| Subjects: | |
| ISBN: | 9781450343916, 1450343910 |
| Online Access: | Get full text |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | In his 1986 Automatica paper Willems introduced the influential behavioural approach to control theory with an investigation of linear time-invariant (LTI) discrete dynamical systems. The behavioural approach places open systems at its centre, modelling by tearing, zooming, and linking. We show that these ideas are naturally expressed in the language of symmetric monoidal categories.
Our main result gives an intuitive sound and fully complete string diagram algebra for reasoning about LTI systems. These string diagrams are closely related to the classical notion of signal flow graphs, endowed with semantics as multi-input multi-output transducers that process discrete streams with an infinite past as well as an infinite future. At the categorical level, the algebraic characterisation is that of the prop of corelations.
Using this framework, we derive a novel structural characterisation of controllability, and consequently provide a methodology for analysing controllability of networked and interconnected systems. We argue that this has the potential of providing elegant, simple, and efficient solutions to problems arising in the analysis of systems over networks, a vibrant research area at the crossing of control theory and computer science. |
|---|---|
| ISBN: | 9781450343916 1450343910 |
| DOI: | 10.1145/2933575.2934556 |