A Dual Gradient Projection Quadratic Programming Algorithm Tailored for Model Predictive Control

• The objective of this work is to derive a QPalgorithm tailored for MPC. More specific, the primary targetapplication is MPC for discrete-time hybrid systems. A desiredproperty of the algorithm is that warm starts should be possibleto perform efficiently. This property is very important for online linear MPC, and it is crucial in branch and bound forhybrid MPC. In this paper, a dual active set-like QP methodwas chosen because of its warm start properties. A drawbackwith classical active set methods is that they often requiremany iterations in order to find the active set in optimum.Gradient projection methods are methods known to be ableto identify this active set very fast and such a method wastherefore chosen in this work. The gradient projection methodwas applied to the dual QP problem and it was tailored for theMPC application. Results from numerical experiments indicatethat the performance of the new algorithm is very good, bothfor linear MPC as well as for hybrid MPC. It is also noticed thatthe number of QP iterations is significantly reduced compared to classical active set methods.

Ämnesord

TEKNIK OCH TEKNOLOGIER  -- Elektroteknik och elektronik -- Reglerteknik (hsv//swe)

ENGINEERING AND TECHNOLOGY  -- Electrical Engineering, Electronic Engineering, Information Engineering -- Control Engineering (hsv//eng)

Nyckelord

Model predictive control

Hybrid systems

Quadratic programming

Mixed integer quadratic programming

TECHNOLOGY

TEKNIKVETENSKAP

Publikations- och innehållstyp

vet (ämneskategori)

kon (ämneskategori)