Nearest stable system using successive convex approximations

Stability is a crucial property in the study of dynamical systems. We focus on the problem of enforcing the stability of a system a posteriori. The system can be a matrix or a polynomial either in continuous-time or in discrete-time. We present an algorithm that constructs a sequence of successive stable iterates that tend to a nearby stable approximation X of a given system A . The stable iterates are obtained by projecting A onto the convex approximations of the set of stable systems. Some possible applications for this method are correcting the error arising from some noise in system identification and a possible solver for bilinear matrix inequalities based on convex approximations. In the case of polynomials, a fair complexity is achieved by finding a closed form solution to first order optimality conditions.