Some methods for reducing the computational complexity of MPC
Event details
| Date | 28.03.2014 |
| Hour | 10:15 |
| Speaker | Prof. Daniel Axehill |
| Location | |
| Category | Conferences - Seminars |
The aim with this talk is to give an overview of some concepts related to efficient computations for MPC. The talk will introduce the basic concepts behind active-set solvers and how such algorithms can be improved in order to reduce the number of iterations necessary. Apart from decreasing the number of iterations, it is interesting to reduce the computational cost for each iteration and it is outlined how a Riccati factorization can be used to reduce this cost. Beyond these standard results, it is shown how structure exploiting rank-k-updates of the Riccati factorization can be performed, how relevant search directions can be computed using parallel computations, and how problem structure still can be exploited for a condensed MPC formulation
Bio: Daniel Axehill works as an assistant professor (forskarassistent) at the Division of Automatic Control at the Department of Electrical Engineering at Linköping University. He started his graduate studies in February 2003 after completing his M.Sc. degree in Applied Physics and Electrical Engineering (Y) at Linköping University. Daniel received his Lic.Eng. degree in December 2005. The Licentiate's thesis can be found here. Daniel received his PhD degree in February 2008. The thesis is available from here. Daniel held a post-doc position at the Automatic Control Laboratory at ETH Zurich from January 2009 until November 2010.
Bio: Daniel Axehill works as an assistant professor (forskarassistent) at the Division of Automatic Control at the Department of Electrical Engineering at Linköping University. He started his graduate studies in February 2003 after completing his M.Sc. degree in Applied Physics and Electrical Engineering (Y) at Linköping University. Daniel received his Lic.Eng. degree in December 2005. The Licentiate's thesis can be found here. Daniel received his PhD degree in February 2008. The thesis is available from here. Daniel held a post-doc position at the Automatic Control Laboratory at ETH Zurich from January 2009 until November 2010.
Links
Practical information
- Expert
- Free