PhD in Safe AI-based Control
PhD opportunity at
Eindhoven University of Technology (TU/e)
in
Safe AI-based Control
, focused on developing neural-network-based controllers that are both expressive and certifiably safe. The project sits at the intersection of
machine learning
and
control theory
, and is designed for a student who wants to contribute to the foundations of trustworthy artificial intelligence for safety-critical applications.
You will investigate
analysis, verification, and design methods
for neural controllers, moving beyond post-training validation to
safety- and verification-by-design
. The work includes rigorous theory and scalable computation for certifying closed-loop safety and stability. Possible directions include
control barrier functions
,
Lyapunov functions
,
quadratic constraints
,
semidefinite programming
, and neural architectures with intrinsic stability or robustness properties. The project also emphasizes combining physical insight, prior system knowledge, and stabilizing baseline controllers with learning to improve performance while maintaining guarantees.
The research will be carried out in the
Control Systems Technology group
in the
Department of Mechanical Engineering
. The group is described as open and collaborative, with expertise in data-driven modelling of dynamical systems, control of complex and uncertain systems, motion control for high-tech systems, model predictive control, networked and supervisory control, neuromorphic and learning-based control, cyber-physical systems, and optimization. The work is supported by close collaboration with industry and will be evaluated on benchmark problems as well as more realistic complex dynamical systems.
The position offers close scientific supervision, freedom to shape your research direction, opportunities to present at international conferences, and the chance to grow as an independent researcher. The broader application areas mentioned include
robotics
,
motion control
, and
energy systems
.
Eligibility highlights:
a master’s degree or equivalent in Systems and Control, Control Engineering, Engineering Cybernetics, Mechanical Engineering, Electrical Engineering, Computer Science, Mechatronics, Artificial Intelligence, or a closely related field; strong interest in control theory, machine learning, and safety-critical systems; a mathematically rigorous and research-oriented mindset; ability to work independently and collaborate across disciplines; motivation to develop teaching skills; and fluent English at C1 level.
Funding and employment:
this is a full-time PhD employment for four years, with an intermediate assessment after nine months. The salary is on Dutch university scale P, with a gross monthly base salary between €3,204 and €4,051, plus an 8% holiday allowance and an 8.3% year-end bonus. TU/e also lists pension contributions, training support, commuting/home-work/internet allowances, and other employee benefits.
Application window:
published 2026-07-29; last application date 2026-08-30. Apply online only using the application link. The application should include a cover letter, CV with publications, and contact information for three references. The vacancy remains open until the position is filled.
