Friday 18 January 2019 13.00 – 16.00, The Technical University of Denmark, Building 101, room S10.\n

\n**Supervisor:** Associate Professor John Bagterp Jørgensen, DTU Compute\n

\n**Co-supervisor:** Professor Henrik Madsen, DTU Compute\n

\n**Co-supervisor:** Associate Professor Niels Kjølstad Poulsen, DTU Compute\n

\n**Co-supervisor:** Associate Professor Bjarne Poulsen, DTU Compute\n

**Examiners:** Professor Michael Pedersen, DTU Compute\n

\nProfessor Carsten W. Scherer, University of Stuttgart, Germany\n

\nAssociate Professor Mark Cannon, University of Oxford, United Kingdom\n

Moderator: Associate Professor Martin Skovgaard Andersen, DTU Compute\n

\n**Summary:\n**

\nIn daily life we are\nsurrounded by an increasing number of devices that each involve some element of automatic\ncontrol.\nDrones and self-driving cars are just the most recent examples.\n

\nIn case a model is available for the dynamics of the system to be controlled, an optimization-based predictive\ncontrol technique may well be an attractive option. However, the faster the systems dynamics, the more stringent\nrequirements it puts on the execution of the optimization. This thesis offers an efficient implementation of an\noptimization problem occurring in the context of linear stochastic systems subject to constraints.\nModelling of the linear system is done in continuous time and comprises both a deterministic and stochastic\nmodel part. To obtain a parsimonious parametrization, transfer functions are used. Also the cost function of the\ncontrol problem is formulated in continuous time. This thesis analyzes the behaviour of the corresponding\ndiscretized control problems for increasingly fast sampling rate. Convergence theorems are proved for the case\nof uniform sampling with sampling time tending to zero.\n

\nOften a single deterministic model is not sufficient to capture the behavior of a system. The description may then\nbe augmented with a stochastic part quantifying the degree of confidence in the deterministic model part. The\npresentation bridges the gap between deterministic and stochastic descriptions for sufficiently regular linear\ntime-variant systems on state space form. It is shown how a description in terms of stochastic differential\nequations (SDE) results if one requires the state equations to hold in the sense of distributions. In particular it\nbecomes clear why the stochastic integral appearing in the SDE must be defined the way it is.

READ MORE about this thesis in DTU Orbit.

\nURL:https://www.compute.dtu.dk/english/Kalender/2019/01/PhD-by-Defense-Morten-Hagdrup DTSTAMP:20220124T072700Z UID:{8548156F-E847-4428-B914-1E16C87AE956}-20190118T120000Z-20190118T120000Z LOCATION: DTU, Building 101, room S10 END:VEVENT END:VCALENDAR