Subspace tracking for online system identification

Abstract: This paper introduces an online approach for identifying time-varying subspaces defined by linear dynamical systems, leveraging optimization on the Grassmannian manifold leading to the Grassmannian Recursive Algorithm for Tracking (GREAT) method. The approach of representing linear systems by non-parametric subspace models has received significant interest in the field of data-driven control recently. We view subspaces as points on the Grassmannian manifold, and therefore, tracking is achieved by performing optimization on the manifold. At each time step, a single measurement from the current subspace corrupted by a bounded error is available. The subspace estimate is updated online using Grassmannian gradient descent on a cost function incorporating a window of the most recent data. Under suitable assumptions on the signal-to-noise ratio of the online data and the subspace's rate of change, we establish theoretical guarantees for the resulting algorithm. More specifically, we prove an exponential convergence rate and provide a consistent uncertainty quantification of the estimates in terms of an upper bound on their distance to the true subspace. The applicability of the proposed algorithm is demonstrated by means of numerical examples, and it is shown to compare favorably with competing parametric system identification methods.