NCS Toolbox
Robust Stability Analysis Made Easy
In recent years, there have been many theoretical developments in the area of stability analysis for networked control systems (NCSs), however, computational tools which implement the theory are lacking. Although toolboxes which analyze NCSs currently do not exist, toolboxes which simulate NCSs have existed for almost a decade. Simulation toolboxes are useful because they can provide an accurate snapshot impression of how the NCS will behave, but due to the inherent uncertainties of NCSs, simulations can not provide any (long term) guarantees of control relevant properties. Therefore, analysis toolboxes that guarantee control relevant properties of NCSs are needed to provide control engineers with tools that can bestow. For this reason, we have developed a prototype MATLAB toolbox with the primary goal of increasing the number of people that can interact with and use the stability theory for NCSs in a simple manner.
In fact, there are multiple other reasons for creating an NCS toolbox: (i) to provide a user-friendly way to apply the existing theoretical developments to NCS problems of choice; (ii) to remove the burden of implementing some of the more complex algorithms employed in the theory; (iii) to gain feedback from the community to improve the quality and overall usefulness of both the toolbox and the theory. To appeal to the widest range of people, we have designed the toolbox such that it is useful for both the basic user (general control engineers) and the advanced user (members of the NCS community).
To appeal to the basic user, we have completely automated the stability analysis procedure such that the user is only required to input a plant model, a controller model, and bounds on the network uncertainties and can directly start analyzing robust stability of the corresponding closed-loop NCS. Moreover, we have recognized that NCS models can contain a considerable amount of variables, which can be intimidating to someone who is not familiar with the NCS nomenclature.
To appeal to the advanced user, we enabled the use of (command-line) functionality outside the GUI to aid in analyzing more speci c problem settings. First, researchers can add functionality for stability analysis to the toolbox by writing their own programs which can visualize entire stability regions for their problem of interest. Second, to encourage community participation in the future development of the toolbox, we made it possible for the researcher to incorporate their own discrete-time models.