Sparse preconditioning for model predictive control
arXiv:1512.00375 · doi:10.1109/ACC.2016.7526060
Abstract
We propose fast O(N) preconditioning, where N is the number of gridpoints on the prediction horizon, for iterative solution of (non)-linear systems appearing in model predictive control methods such as forward-difference Newton-Krylov methods. The Continuation/GMRES method for nonlinear model predictive control, suggested by T. Ohtsuka in 2004, is a specific application of the Newton-Krylov method, which uses the GMRES iterative algorithm to solve a forward difference approximation of the optimality equations on every time step.
6 pages, 5 figures, to appear in proceedings of the American Control Conference 2016, July 6-8, Boston, MA, USA. arXiv admin note: text overlap with arXiv:1509.02861