Pavel Otta: On the Customizable QP Formulation for MPC

Fri, 06/01/2018

Starting at 10am in the K14 seminar room, Pavel Otta - doctoral student of Vladimir Havlena's - is going to give us an update on his recent achievements.

Abstract:  This seminar will addrese a systematic solution of Model Predictive Control (MPC) problem. We start with favorite dense and sparse Quadratic Programming (QP) formulations of the problem. We will discuss the suitability of each other and the connection between these two. Then we recall partial-condensing and move-blocking concepts and show the combination the two previous serves a customizable QP formulation of the MPC problem. This seminar does not require any advanced mathematics skills, but only simple matrix computation. Hopefully, everyone experiences an Aha! moment from the connection of basic concepts.