Automatic code generation for linear model predictive control
Implementing model predictive control (MPC) schemes into embedded platforms
for systems with fast dynamics (e.g. mechatronic systems) is a hot research
topic. Standard general purpose MPC algorithms are not appropriate for such
platforms, due to the high memory and computational requirements. At our
institute we have developed a free automatic code generation software called
μAO-MPC.
However, at the moment
μAO-MPC can only handle a very particular
type of linear time-invariant
discrete-time systems (the
ltidt module).
The proposed project aims to extend
μAO-MPC's capabilities by
developing a new module that can generate code for a broader range of
linear discrete-time systems.
This rather challenging task offers the
possibility of writing a scientific publication afterwards. The written
language of the project is English. Therefore, good command of written
English is required.
The programming language is Python and C (previous knowledge is not required).
We are looking for highly motivated
Master students
with some programming experience.
If you are interested in this topic, please discuss the details with
the
contact
person.
Area:
Convex optimization, optimal control,
automatic code generation,
fast model predictive control algorithm.
Helpful/Required Prerequisites:
Lectures: Regelungstechnik, Optimal Control.
Software: Python and C programming languages (previous experience not required).
Language: English.
Start:
As soon as possible
Requirements:
| Literature search |
40% |
| Programming |
50% |
| Experimental work |
10% |
Pablo Zometa