The Delft Center for Systems and Control ([https://www.dcsc.tudelft.nl]) of Delft University of Technology, The Netherlands has a vacancy for 3 PhD positions on "Integrated optimization-based and learning-based control of networks with hybrid dynamics".
These 3 PhD projects are part of the European ERC Advanced Grant project CLariNet – a novel control paradigm for large-scale hybrid
networks. The goal of CLariNet is to create a completely new paradigm for control of large-scale networks with hybrid dynamics by bridging
the gap between optimization-based control and learning-based control. The breakthrough idea is to bridge that gap by using
piecewise affine models and to unite the optimality of optimization-based` control with the on-line tractability of learning-based control.
The 3 PhD projects all have a strong fundamental flavor. In addition, applications for the case studies include multi-modal transportation
networks and smart multi-energy networks.
Topic 1: Learning-based control for hybrid systems with constraints
In this PhD project we will develop learning-based control methods for hybrid systems — in particular piecewise affine (PWA) systems — that
allow to include explicit linear or convex, mixed constraints on the inputs, states, and outputs of the PWA system. Subsequently, the
approach will be extended to a multi-agent setting using a combination of optimization-based and learning-based control for large-scale
networks with PWA dynamics, with emphasis on the development of numerically reliable and computationally efficient algorithms.
Topic 2: Integrated optimization-based and learning-based control for hybrid systems
In this PhD project we will develop integrated optimization-based and learning-based control methods for hybrid systems — in particular
piecewise affine (PWA) systems. More specifically, the aim is to develop several innovative approaches to combine model predictive
control (MPC) and reinforcement learning so as to merge the advantages of both approaches.
Topic 3: Multi-scale multi-resolution models for large-scale networks with hybrid dynamics
In this PhD project we will develop a framework for multi-agent integrated optimization- based and learning-based control of
large-scale networks. More specifically, we address two challenges: (1) developing methods to determine spatial and temporal divisions
that are appropriate for the proposed control structure, and (2) developing appropriate piecewise-affine models for use within the
multi-agent control framework that allow different levels of spatial and temporal modeling detail according to the requirements, i.e.,
multi-scale multi-resolution models.
What are we looking for?
We are looking for a candidate with an MSc degree in systems and control, applied mathematics, computer science, electrical
engineering, or a related field, and with a strong background or interest in: machine learning, optimization, and control (for Topic
1); systems & control, machine learning, and optimization (for Topic 2); distributed control and modeling (for Topic 3). The candidate is
expected to work on the boundary of several research domains. A good command of the English language is required.
What do we offer?
We offer the opportunity to do scientifically challenging research in a multi-disciplinary research group. The appointment will be for up to
4 years. The PhD student will also be able to participate in the research school DISC ([http://www.disc.tudelft.nl]). As an employee of
the university you will receive a competitive salary, as well as excellent secondary benefits. Assistance with accommodation can be
arranged.
How to apply?
More information on this position and on how to apply can be found at [https://www.dcsc.tudelft.nl/~bdeschutter/vac/vacancy_phd_1_2_3_clarinet.pdf] or by contacting Bart De Schutter (b.deschutter at tudelft.nl).
