Abstract:
Recently, the novel framework for data-driven analysis and control, called data informativity, was proposed. This notion represents whether the given data contain suffici...Show MoreMetadata
Abstract:
Recently, the novel framework for data-driven analysis and control, called data informativity, was proposed. This notion represents whether the given data contain sufficient information to solve a problem or not. This letter characterizes the data informativity for Lyapunov equations in the form of AP + PA^{\top } = -Q . First, we clarify the relationship between the unique solution and an invariant subspace of the matrix A . Second, based on this result, we provide a necessary and sufficient condition for the data informativity, which is characterized by the possibility of the matrix decomposition of Q . Finally, we present a data-driven method for solving the Lyapunov equation.
Published in: IEEE Control Systems Letters ( Volume: 7)