Automatic Control Laboratory

Open Opportunities

Advanced Volume Control for Pipetting ETH Zurich Automatic Control Laboratory Improving volume control precision and robustness in automated pipetting remains a challenge, often limited by traditional indirect methods. This project explores direct volume control by leveraging internal air pressure measurements and the ideal gas law. Key obstacles include friction, pressure oscillations, varying liquid viscosities, evaporation, and liquid retention. Collaborating with Hamilton Robotics, the goal is to develop a robust control architecture for their precision pipette (MagPip) suitable for diverse liquids. The approach involves mathematical modeling based on sensor data, designing robust control strategies to handle nonlinearities and disturbances, and validating through simulation and real-world experiments. Control Engineering, Systems Theory and Control, Systems Theory and Control Semester Project

Feedback Optimization for Freeway Ramp Metering ETH Zurich Automatic Control Laboratory Online Feedback optimization (OFO) is a beautiful control method to drive a dynamical system to an optimal steady-state. By directly interconnecting optimization algorithms with real-time system measurements, OFO guarantees robustness and efficient operation, yet without requiring exact knowledge of the system model. The goal of this project is to develop faster OFO schemes for congestion control on freeways, in particular by leveraging the monotonicity properties of traffic networks. Engineering and Technology Master Thesis

Data-driven Safe Control Design: A Certificate Function Approach ETH Zurich Automatic Control Laboratory Safety is a fundamental requirement for critical systems such as power converter protection, robotics, and autonomous vehicles. Ensuring long-term safety in these systems requires both characterizing safe behaviour and designing feedback controllers that enforce safety constraints. Control Barrier Functions (CBFs) have recently emerged as a powerful tool for addressing these challenges by defining safe regions in the state space and formulating control strategies that maintain safety. When the dynamical system is precisely modeled, a CBF can be designed by solving a convex optimization problem, where the state-space model is incorporated into the constraints. However, designing valid CBFs remains difficult when system models are uncertain or time-varying. More importantly, CBFs and control laws derived from inaccurate models may lead to unsafe behaviour in real-world systems. To overcome these difficulties, this project aims to develop a data-driven approach for constructing CBFs without relying on explicit system models. Instead, we will leverage behavioural systems theory to replace model information in the design program by persistently exciting data. The proposed method will be applied to output current protection in power converters or robotics collision avoidance. Engineering and Technology Master Thesis, Semester Project

Data-Driven Power System Stabilization with Dissipativity-Informed Neural Networks ETH Zurich Automatic Control Laboratory Modern power systems exhibit significant complexity, making their analysis and control particularly challenging, especially when precise system models are unavailable. Traditional model-based control strategies often fail to scale with increasing system complexity, while recent advances in nonlinear, learning based control offer promising alternatives. However, many of these methods lack formal stability guarantees, which are crucial for safety-critical applications such as power system frequency control. This project aims to bridge this gap by developing a deep learning framework for analyzing the dissipativity properties of power systems and designing stabilizing controllers with formal guarantees. Engineering and Technology Master Thesis, Semester Project

Designing High-Performance MPC Controllers Under Environmental Changes Using Meta-Learning ETH Zurich Automatic Control Laboratory Model predictive control (MPC) is a widely used control technique that optimizes control inputs while fulfilling process constraints. Although automated tuning methods have been developed for task-specific MPC, they struggle when tasks change over time, requiring costly re-tuning. This project aims to reduce the computational burden of re-tuning by leveraging meta-learning, enabling efficient adaptation of controllers to different environments with minimal data. Electrical Engineering Semester Project

Identifying influencers in social networks ETH Zurich Automatic Control Laboratory The objective of this project is the design and analysis of recommender systems as optimization algorithms representing a robust feedback controller. We aim to design recommender system algorithms that identify influential users using observable data from users (for example: clicks/ time spent on a page/ likes etc.) in a social network and provide recommendations accordingly. Engineering and Technology, Mathematical Sciences Master Thesis, Semester Project