Scalable Control Approximations for Resource Constrained Environments

Introduction

This project aims at making a breakthrough contribution in optimal control and decision making for nonlinear processes that take place on network structures and are dynamic in time and/or space. The setting has a wide range of potential domains of applicability, comprising thermal, electric, or fluid dynamics in energy networks, logistics, disease spreading dynamics, or cell signalling in biomedicine.

Objectives

The project will pursue the following objectives:

1. To contribute new theory.
2. To develop numerical approximation methods.
3. To implement algorithmic methods in software.
4. To conduct proof-of-concept studies.

Background

Research in the young field of mixed-integer optimal control (MIOC) has recently seen increased momentum together with numerical approximation algorithms and control theory. Despite initial successes, key questions remain unsolved because of:

• A lack of analytical understanding.
• A lack of tractable formulations.
• The unavailability of efficient solvers.
• The insufficiency of existing implementations.

Focus Areas

This project focuses on pivotal but poorly understood topics:

• Decomposition, relaxation, and approximation.
• Domains admitting homogenization and limiting processes using weak topologies.
• Tractable approximations of direct costs of decisions.
• Efficient distributed and parallel nonlinear solvers.
• Robustness of approximate nonlinear decision policies under uncertainty.

These key issues appear systematically in a wide range of control tasks of high societal relevance.

Contribution to the Field

By addressing them, the project helps to bridge a persistent and pronounced gap in simulation & optimization practice. Due to non-trivial interactions emerging in theory and the unavailability of comprehensive algorithms, these topics cannot be suitably handled by merely combining the respective states of the art.

A focused effort to decisively extend MIOC to optimal decisions for dynamics on networks is therefore a timely endeavour that will help to address the challenging demands of practitioners.