Industrial Applications for Embedded Optimization

Stefan Gering

Robert Bosch GmbH, Renningen, Control Theory (CR/AEH3)

Tuesday, December 20, 2016, 10:00

Room 02-016, Georges-Koehler-Allee 101, Freiburg 79110, Germany

Starting from first process engineering applications, model predictive control has up until now been in the focus of research and development for some decades. Advantages of this control method based on the receding horizon principle are well known and widely documented in the literature. Besides further development of embedded optimization algorithms towards computational efficiency, one remaining challenge is how to incorporate such algorithms in actual products, for which model predictive control is not only a modern substitute to classical PID control, but a core technology allowing to exploit system degrees of freedom in an optimal manner.

This talk will focus on the BOSCH perspective on model predictive control and its potential application fields in the automotive environment. First, an overview on corporate research activities is given emphasizing business fields and potential products, in which MPC may be applied to. Second, a variety of applications within the automotive sector is focused on allowing for different MPC strategies with different levels of computational demand and constraints.