< Back to previous page

Project

Risk-averse Control and Dynamic Optimization: Bridging the Gap Between Robust and Stochastic Control

In this research project a novel framework for risk averse model predictive control (MPC) will be developed, as existing techniques like robust MPC are too conservative or in the case of stochastic MPC require knowledge on the uncertainty distribution. Model predictive control is an online optimization-based control scheme which was originally developed for chemical processes but nowadays has found applications in a wide range of fields, such as robotics, (semi-)autonomous vehicles, artificial pancreas and many another. As the main focus of this project is on the development of novel control theory and corresponding optimization methods up to their experimental validation, the proposers believe that the most suitable expert panel is "W&T7: Energy, Electrical Engineering, Electronics and Mechanics" (key-word: control engineering). Besides the development of novel control theory, data-driven approaches for system (re)identification in view of the design of risk-averse controllers, constitute a substantial research effort within this proposal. This is an additional motivation for the proposers to situate the research project within the chosen expert panel (keyword: system identification and control). In addition there is a smaller research task related to computational speed-up by exploiting Graphical Process Unit (GPU) which is an additional reason to situate the proposal in W&T7 (keyword: micro- and nano- electronics).

Date:1 Jan 2018 →  31 Dec 2021
Keywords:Control engineering, System identification and control, Micro-electronics, Nano-electronics, Model predictice control
Disciplines:Control systems, robotics and automation, Design theories and methods, Mechatronics and robotics, Computer theory