determine relations between multivariable dynamic models in form of state space models and transfer functions
analyse multivariable dynamic systems with respect to stability, sensitivity for disturbances, statistical properties, and controllability and observability
analyse dynamic systems influenced by noise, and to determine stationary variances for given linear models
design optimal observers (Kalman filters)
design controllers for linear multivariable systems based on linear quadratic (LQ) control
account for the principles behind model predictive control (MPC)
evaluate controllers in laboratory work on real processes
Mathematical description of linear multivariable systems in continuous and discrete time. Controllability and observability. Stability. Description of disturbances and their effects. Controller synthesis using linear quadratic theory and the separation theorem. Model predictive control.
Lectures, problem solving sessions, tutorials and laboratory work. Guest lecture. Non-compulsory homework assignments.
Written examination at the end of the course (4 credits). Passed laboratory course is also required (1 credit).