TR2016-046

Sparse Preconditioning for Model Predictive Control


    •  Knyazev, A.; Malyshev, A., "Sparse Preconditioning for Model Predictive Control", American Control Conference (ACC), DOI: 10.1109/ACC.2016.7526060, July 2016, pp. 4494-4499.
      BibTeX Download PDF
      • @inproceedings{Knyazev2016jul,
      • author = {Knyazev, A. and Malyshev, A.},
      • title = {Sparse Preconditioning for Model Predictive Control},
      • booktitle = {American Control Conference (ACC)},
      • year = 2016,
      • pages = {4494--4499},
      • month = jul,
      • doi = {10.1109/ACC.2016.7526060},
      • url = {http://www.merl.com/publications/TR2016-046}
      • }
  • Research Area:

    Control


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.