Stabilisation of large-scale nonlinear systems by modifying the interconnection network

In this article, we develop a new strategy for the robust stabilisation of large-scale singular systems. The models that we will be concerned with consist of a large number of nonlinear subsystems, which are linked through a linear interconnection network. This approach does not require the availability of control inputs, and focuses instead on variable parameters in the interconnection network. It is shown that the determination of appropriate parameter values reduces to a constrained static output feedback problem, which can be solved using linear matrix inequalities. The proposed method is designed to minimise the number of optimisation variables, and is therefore suitable for large-scale applications. It can also incorporate information structure constraints and uncertainties in the system model.