动态系统中学习算子控制理论方向博士研究员

Are you fascinated by the intersection of control theory and scientific machine learning? Join us in developing the mathematical foundations needed to make deep operators reliable, robust, and applicable for control of complex engineering systems. In this PhD project, you will investigate how operator-learning models can represent dynamical systems, how their stability and robustness can be characterized, and how controllers can be designed for them. Your work will combine rigorous theory, numerical methods, and applications in high-tech, medical, and energy systems.

Information Modern engineering increasingly relies on data-driven models to describe complex dynamical systems. Scientific machine learning is now enabling a new class of models that learn operators mapping system inputs and initial conditions directly to system responses. These learned operators offer exciting opportunities for efficient simulation, digital twins, optimization, and control.

In this PhD project, you will develop a new systems and control theory for learned operators, bridging modern scientific machine learning with classical control theory. Rather than focusing on a single machine-learning architecture, you will establish mathematical principles that apply across a broad class of operator-learning methods, including current and future deep operator models.

Your research will investigate fundamental questions such as: when does a learned operator admit an equivalent state-space representation? How can stability, contraction, dissipativity, robustness, and other system-theoretic properties be defined and guaranteed directly in the operator domain? And how can feedforward, feedback, and predictive controllers be designed for systems represented by learned operators? You will combine analytical theory with numerical implementation and validate the developed methods on industrially relevant applications in the high-tech, medical, and energy sectors.

You will become part of the Dynamics and Control group at the Department of Mechanical Engineering at Eindhoven University of Technology, under the supervision of dr.ir. Fahim Shakib and prof.dr.ir. Nathan van de Wouw. You will contribute to research at the intersection of systems and control theory, scientific machine learning, operator theory, optimization, and applied mathematics. You will work in an interdisciplinary environment with opportunities to collaborate with academic and industrial partners. You will be part of a vibrant and socially interactive group of approximately 35 researchers in the field of dynamical systems and control.