Sections

Robust nonlinear Model Predictive Control (Reference: ESR10) **** This position has been filled ****

Location: University of Bayreuth (application deadline: November 30, 2010 or until position filled). *** Provided position ***

  • Title: Robust nonlinear Model Predictive Control
  • Location:  University of Bayreuth, Germany
  • Advisor: Lars Grüne and Matthias Gerdts
  • Secondment: Imperial College, London, UK
  • Contact: Lars Grüne
  • Reference: ESR10

  • Key Words: model predictive control, robustness, sensitivity, numerical optimal control

  • Application deadline: November 30, 2010, will be extended until the position is filled

  • Description of the subject: Model predictive control (MPC) is an approach to control feedback design, based on the solution, at each controller update step, of an optimal control problem. This approach has shown to be a viable approach to nonlinear optimal control. However, when MPC faces uncertainties and disturbances, several difficulties arise, and in practice, only suboptimal robust control strategies can be achieved. In the framework of this project we investigate aspects of robust nonlinear MPC, including stability and suboptimality estimates, and develop new MPC algorithms that take account of the limited time, between control update times, available for computation. Particularly, the use of efficient numerical methods based on sensitivity methods for counteracting the effect of disturbances shall be investigated and implemented.

  • Requirements: The candidate for the PhD position in this project should preferably have experience in one or more of the following areas:

    - nonlinear model predictive control
    - numerical methods for optimal control, including sensitivity analysis
    - stability analysis of nonlinear systems
    - programming experience
Document Actions