SA/BA/MA projects

Have you heard of industry 4.0, smart grids, smart buildings, smart cities, intelligent traffic systems and intelligent self-driving cars? Did you ever wonder what makes them smart and intelligent?

The answer is control and automation and that’s what we do at the Automatic Control Laboratory (IfA). We use control theory, optimization, machine learning, and game theory to develop controllers and algorithms that are the backbone of nearly all modern technology. At our lab we span the whole area from pure theory to real-world applications and we are looking for you to help us push forward the state of the art. If you want to learn techniques and gain knowledge that enable you to work in any field from medical applications to spacecraft and from electrical grids to finance then the Automatic Control Laboratory is the place for you!

Lists of currently running and recently completed projects can be found in the sub-navigation menu above. Reports for several projects are available through the ETHZ Research Collection:

Open projects

Experimental Validation of a Modeling Method for Impedance Identification in Three-Phase Power Systems

This project aims to use two converter emulators available in the Automatic Control Laboratory of ETHz to experimentally validate a new impedance estimation approach. The main goals are to replicate realistic converter/grid conditions, assess the accuracy and robustness of the estimation method, and to explore its limitations and performance boundaries.

Keywords

Impedance Estimation; Grid-connected converters; Power electronics; System Identification

Labels

Master Thesis

Description

The integration of renewable energy sources and distributed generation in modern power systems has increased significantly. This resulted in an increased use of power-electronics-based converters, changing the dynamical behaviour of the grid and raising the need for accurate grid impedance estimation methods. Knowledge of grid impedance at the point of common coupling is essential for online stability assessment of grid-connected converters [1]. It is also used for optimizing the closed-loop performance via, e.g., adaptive control design. However, due to the grid's dynamic and nonlinear nature, impedance estimation becomes particularly challenging, in particular when conducted in real-time under closed-loop operation (see e.g., [2]). A new non-parametric identification method has shown potential for accurate and rapid grid impedance identification. While initial simulations have demonstrated the potential of the method, experimental validation using hardware is crucial to evaluate the method's performance under real-world conditions. This project aims to use two converter emulators ([3]) available in the Automatic Control Laboratory to experimentally validate the new approach. The main goals are to replicate realistic converter/grid conditions, assess the accuracy and robustness of the method, and to explore the limitations and performance boundaries of the method. References: [1] J. Sun, "Impedance-Based Stability Criterion for Grid-Connected Inverters," in IEEE Transactions on Power Electronics, vol. 26, no. 11, Nov. 2011 [2] V. Haberle, et.al., "MIMO Grid Impedance Identification of Three-Phase Power Systems: Parametric vs. Nonparametric Approaches," 2023 62nd IEEE CDC [3] https://imperix.com/products/power-electronics-test-bench/ **Prerequisites** We are looking for a motivated student with a background in power systems/power electronics, and a previous exposure to system identification or signal processing. Knowledge of grid-connected converters and impedance estimation methods would be highly beneficial. Familiarity with experimental configurations and lab instrumentation is desirable. Competitive candidates will have prior research or work experience that demonstrate the ability to learn and problem-solve independently.

Goal

Contact Details

More information

Open this project...

Published since: 2024-11-15 , Earliest start: 2024-11-17

Applications limited to ETH Zurich

Organization Automatic Control Laboratory

Hosts Abdalmoaty Mohamed , He Xiuqiang

Topics Engineering and Technology

Optimal Excitation for Grid Impedance Estimation

This project aims to develop optimal excitation schemes for impedance estimation of grid/grid-connected converters.

Keywords

Impedance estimation; grid-connected converters; optimal excitation; experiment design; system identification

Labels

Master Thesis

Description

The use of renewable energy sources and distributed generation, such as wind turbines and solar power plants, in modern power systems has been increasing significantly. This resulted in a heightened use of power-electronics-based converters, changing the dynamical behaviour of the power grid and raising the need for accurate grid impedance estimation methods. Knowledge of grid impedance at the point of common coupling is essential for online stability assessment of grid-connected converters [1]. However, due to the grid's dynamic and nonlinear nature, impedance estimation is particularly challenging (see e.g., [2]). The grid-connected converter's controller must maintain stable operation while simultaneously allowing the injection of small excitation signals necessary for grid impedance estimation. This creates a fundamental trade-off: while a good controller and converter-grid interface filter improve system stability and disturbance rejection, they can also suppress the grid's response to the excitation, reducing the estimation accuracy. On the other hand, injecting larger excitation signals could improve the quality of estimation but risks triggering the converter protection system and/or exciting nonlinear grid behaviors. Moreover, it is not clear what the best class of excitation signals is nor the best injection point within the converter controller. This project aims to address this trade-off and questions by developing optimal excitation schemes for grid impedance estimation using grid-connected converters. A main goal is to obtain a good signal-to-noise ratio at the frequencies of interest while respecting experimental time and signal magnitude constraints. A class of excitation signals of potential interest is random phase multisines [3]. Recent related results in the literature are [4,5] which are based on modelling the noise and the use of multiple sinusoidal injections. References: [1] J. Sun, "Impedance-Based Stability Criterion for Grid-Connected Inverters," in IEEE Transactions on Power Electronics, vol. 26, no. 11, Nov. 2011 [2] V. Haberle, et.al., "MIMO Grid Impedance Identification of Three-Phase Power Systems: Parametric vs. Nonparametric Approaches," 2023 62nd IEEE CDC [3] Godfrey, Keith, ed. Perturbation signals for system identification. Prentice Hall International, 1993. [4] Y. Zhu, et.al, "Injection Amplitude Guidance for Impedance Measurement in Power Systems," in IEEE Transactions on Power Electronics, vol. 38, no. 6, June 2023 [5] H. Li, et.al., "A Noise Interference Suppression Method With Optimized Frequency Adjustment for Impedance Measurement," in IEEE Transactions on Industrial Electronics, vol. 71, no. 12, Dec. 2024 **Prerequisits** We are looking for a motivated student with excellent written and verbal communication skills, and the ability to learn and problem-solve independently. Competitive candidate will have a background in systems & control, signal processing, and power systems/electronics. A prior experience or coursework in system identification and/or grid-connected converters would be highly beneficial. Proficiency in MATLAB/Simulink or a similar simulation environment is required for implementing and testing the methods.

Goal

Contact Details

More information

Open this project...

Published since: 2024-11-15 , Earliest start: 2024-11-17

Applications limited to ETH Zurich

Organization Automatic Control Laboratory

Hosts He Xiuqiang , Abdalmoaty Mohamed

Topics Engineering and Technology

Optimal Control of Plants in Hydroponic Systems

This project deals with the optimal control of crops in a hydroponics system. A hydroponics system is a controlled environment in which crops grow in a nutrient solution instead of soil. The goal is to design an algorithm that leverages data to optimally control the environmental conditions of the crop. The objective is to achieve a fast crop growth with as little as possible energy investments.

Keywords

Control Theory, Formal Methods, Agriculture, Hydroponics, Safety, Stochastic Systems, Reinforcement Learning

Labels

Master Thesis

Description

Goal

Contact Details

More information

Open this project...

Published since: 2024-11-15

Applications limited to ETH Zurich

Organization Automatic Control Laboratory

Hosts Wallington Kevin , Schmid Niklas

Topics Mathematical Sciences , Information, Computing and Communication Sciences , Engineering and Technology

Enhancing Model Predictive Control with Reinforcement Learning

Model Predictive Control (MPC) is extensively utilized in industry and academia thanks to its ease of use and flexibility. However, MPC is an inherently suboptimal control technique, and could perform poorly in presence of external disturbances or unmodelled dynamics. Many solutions that aim at robustifying MPC exist, but they are generally overly conservative and difficult to implement. This project seeks to obtain robust MPC schemes that achieve high performance in challenging control tasks by using tools from reinforcement learning through the application of gradient-based optimization schemes.

Keywords

Model predictive control, Reinforcement Learning

Labels

Master Thesis

Description

Model predictive control (MPC) is ubiquitus in industry and academia. The core idea of this control technique is to utilize a model of the process dynamics to construct a prediction of how the system will evolve within a certain prediction horizon, starting from a given initial state. Next, the MPC dynamically determines the optimal control actions within the horizon by solving an optimization problem with the goal of minimizing some cost function (for example steering the state of the system to the origin or tracking a reference). When deployed on a real system, the optimization is repeated at each time-step, and the control sequence is applied only for the first portion of the prediction horizon. As time progresses, the optimization process is repeated, and the horizon "recedes" forward in time. This constant re-optimization ensures that the controller actively responds to variations of the environment, thus enabling feedback. The prediction that the MPC makes at every timestep is referred to as "open-loop" control, since it doesn't take into account future measurements beyond the current time. The main source of suboptimality in MPC is the discrepancy between the open-loop prediction and the true trajectory (closed-loop) obtained with the receding horizon paradigm. This is because the objective function in the MPC problem involves the open-loop, rather than the closed-loop. One way to reduce the suboptimality is to increase the length of the prediction horizon. In this way the MPC is able to capture the system's evolution over a longer period, thereby making choices that better discount between short and long term benefits. However, a longer prediction horizon requires more computational resources and this may not be possible in applications where the MPC is expected to run at fast rates. Another possibility to improve the closed-loop performance of MPC is to modify the cost function of the MPC problem in such a way that the control action is chosen optimally for the closed-loop. We can obtain the optimal cost function by solving a policy optimization problem. These problems, generally addressed in the field of reinforcement learning, involve adjusting the parameters of the policy (in this case, the MPC controller) to improve its performance. Reinforcement learning algorithms can be implemented using gradient-based methods. These methods involve computing the gradient of some performance measure with respect to the MPC parameters and then updating the parameters in the direction that improves performance. To obtain the gradient of the entire closed-loop trajectory with respect to the design parameters, we can utilize the approach proposed in [1], which utilizes a backpropagation scheme. The same gradient-based techniques can also be used to ensure that the MPC scheme is robust against unmodeled dynamics and / or stochastic noise [2] (see pdf for a numerical example).

Goal

Contact Details

More information

Open this project...

Published since: 2024-11-12 , Earliest start: 2024-11-30

Organization Automatic Control Laboratory

Hosts Zuliani Riccardo

Topics Mathematical Sciences , Information, Computing and Communication Sciences

Conducting an Orchestra

Various strategic interactions involve hierarchical decision-making processes, where one entity leads and others react accordingly. Stackelberg games provide a mathematical framework to model such scenarios, capturing the dynamics between a leader and multiple followers. However, in many real-world applications of such structures, we often only observe the response of the followers but we are unsure about the optimization problem that the followers are optimizing. This research question, also known as inverse game theory, poses significant challenges, further complicated by noisy observations, bounded rationality, and many more. This project aims to develop methodologies for inferring the utility functions of followers in such scenarios by leveraging observed actions and partial knowledge of their parameters, working on Swissgrid energy market data provided by the MAESTRO project.

Keywords

Game Theory, Learning, Data Analysis, Energy Market

Labels

Master Thesis

Description

Goal

Contact Details

More information

Open this project...

Published since: 2024-11-11 , Earliest start: 2024-11-17 , Latest end: 2025-07-31

Organization Automatic Control Laboratory

Hosts Shilov Ilia

Topics Mathematical Sciences , Information, Computing and Communication Sciences , Economics

Data-driven Control in Building Energy Systems

Modern buildings' HVAC (Heating, Ventilation, and Air Conditioning) systems incorporate a complex network of sensors, control units, and actuators working in coordination across multiple levels to ensure optimal operation. Key building control tasks include regulating air quality, temperature, and ventilation. Achieving efficient building control is critical for occupant comfort and meeting energy efficiency and sustainability targets. Due to the substantial energy consumption associated with buildings, enhancing operational efficiency by leveraging data analytics for control has a high potential for energy savings and sustainability gains. Effective control strategies can, in many practical cases, significantly reduce CO2 emissions from buildings.

Keywords

Data-Driven Control, Adaptive Control, DeePC, Reinforcement Learning, Buildings, HVAC

Labels

Master Thesis

Description

Goal

Contact Details

More information

Open this project...

Published since: 2024-11-11 , Earliest start: 2024-11-10

Organization Automatic Control Laboratory

Hosts Balta Efe , Spoek Ben

Topics Information, Computing and Communication Sciences , Engineering and Technology

Becoming Ungovernable

This project will investigate how the assumption of rationality affects leader-follower dynamics in Stackelberg games, particularly focusing on the potential loss of the leader’s first-mover advantage when followers act irrationally. We will examine scenarios where followers employ non-credible threats, take into account empirical evidence of irrational behavior and frame communication noise as a form of bounded rationality among followers. The aim of the project is to show that followers can strategically exploit their ”irrationality” to diminish the leader’s influence and to propose new insights into strategic interactions where rationality cannot be assumed, with implications for policy-making and other leader-follower contexts.

Keywords

Game Theory, Mechanism Design, Bounded Rationality, Learning in Games

Labels

Master Thesis

Description

Goal

Contact Details

More information

Open this project...

Published since: 2024-11-11 , Earliest start: 2024-11-17 , Latest end: 2025-07-31

Organization Automatic Control Laboratory

Hosts Shilov Ilia

Topics Mathematical Sciences , Economics

System theory of iterative methods

Modern control methods often rely on explicit online computation. In order to understand such closed loops between numerical methods and dynamical systems, this project approaches the algorithm as a dynamical system itself. In doing so, the usual language of convergence of algorithms can be viewed as a special case of stability theory.

Labels

Master Thesis

Description

Goal

Contact Details

More information

Open this project...

Published since: 2024-10-18

Organization Automatic Control Laboratory

Hosts Eising Jaap

Topics Mathematical Sciences , Information, Computing and Communication Sciences , Engineering and Technology

Data-Driven Adaptive Control for Linear Time-Varying Systems

Time-varying effects in engineering systems, such as wear and unmodeled dynamics, require adaptive controllers that adjust in real time. Traditional methods rely on system identification, while data-driven approaches like data-enabled policy optimization (DeePO) directly optimize controllers using real-time data. DeePO has shown both theoretical gurantees for linear systems and good performance for nonlinear applications. This project aims to extend DeePO to linear time-varying systems in both theory and simulation, where the simulation example is pitch control of unmanned aircraft.

Keywords

Adaptive control, data-driven control, aircraft control.

Labels

Semester Project , Master Thesis , ETH Zurich (ETHZ)

Description

Goal

Contact Details

More information

Open this project...

Published since: 2024-10-15 , Earliest start: 2024-10-28 , Latest end: 2025-07-07

Organization Automatic Control Laboratory

Hosts Zhao Feiran , Eising Jaap

Topics Information, Computing and Communication Sciences , Engineering and Technology

Fast Computation of Dynamic Population Games with Madupite

Dynamic Population Games (DPGs) are an important class of games that models many real-world problems, including energy systems, epidemics, and the recently proposed “karma economies” for fair resource allocation. A DPG consists of a large population of self-interested agents each solving an individual Markov Decision Process (MDP). The MDP of each agent is coupled to the actions of others and is hence parametrized by the policies adopted in the population. Computing the Nash equilibrium of a DPG is challenging as it involves iteratively solving MDPs many times. This suffers from the well-known curse of dimensionality which severely limits the size of the state and action spaces that are computationally tractable. Madupite is a novel distributed high-performance solver for large-scale infinite horizon discounted MDPs, which leverages PETSc to implement inexact policy iteration methods in a distributed fashion. Despite its software complexity, Madupite comes with a very intuitive Python interface and a detailed documentation, that allow any Python user to easily deploy it to efficiently simulate and solve large-scale MDPs in a fully distributed fashion. Preliminary benchmarks have showcased the great potential of Madupite, which is capable of efficiently handling MDPs with millions of states. Motivated by the recent development of Madupite, this project aims at developing fast computation tools that are capable of solving large-scale DPGs.

Keywords

Programming (Python/C++), Markov Decision Processess, Game Theory, Karma Economy

Labels

Semester Project , Bachelor Thesis

Description

Goal

Contact Details

More information

Open this project...

Published since: 2024-09-25 , Earliest start: 2024-09-30

Organization Automatic Control Laboratory

Hosts Elokda Ezzat , Gargiani Matilde

Topics Information, Computing and Communication Sciences , Engineering and Technology

Data-driven control using spectral submanifold for feedback control of nonlinear dynamical systems

In this project, we will explore an alternative approach to learning nonlinear dynamical systems using slow spectral submanifolds (SSMs). SSMs decomposes nonlinear system dynamics into nonlinear eigenspaces, and distinguishes them by their dominant (slow) and transient (fast) behaviors. From these eigenspaces, we can learn nonlinear dynamical models by selecting the dominant eigenspaces in a similar fashion as EDMD-based model approximation.

Keywords

dynamic mode decomposition, data-driven control, spectral submanifolds

Labels

ETH Zurich (ETHZ)

Description

Goal

Contact Details

More information

Open this project...

Published since: 2024-08-20 , Earliest start: 2024-08-20

Organization Automatic Control Laboratory

Hosts Li Hui

Topics Mathematical Sciences

Direct data-driven predictive control for water storage reservoirs

Water storage reservoirs are critical infrastructure for energy production, water supply, and flood protection. The state-of-the-art for operating reservoirs is forecasted informed model predictive control. This project proposes an alternative, data-driven approach - rather than attempting to model the complex dynamics between weather forecasts and reservoir river inflow, the data-driven approach learns these dynamics from data. This thesis seeks to make a notable contribution to data-driven reservoir management.

Keywords

Water resources systems; reservoir operation; data-driven predictive control; optimal control

Labels

Master Thesis

Description

Water storage reservoirs are critical infrastructure for energy production (hydropower), water supply, and flood protection. The uncertainty of meteorological inputs is a major challenge to operating reservoirs to meet these (often conflicting) objectives. However, meteorological forecast skill has improved dramatically in recent decades, and forecast-informed reservoir operation (FIRO) is a promising approach to maximize reservoir benefits and avoid tradeoffs between conflicting objectives. The state-of-the-art for FIRO is a model predictive control (MPC) approach (see Figure 1, pdf attached): (1) forecasted meteorology is fed into a hydrological model that generates a forecasted river inflow to the reservoir; (2) reservoir operations are optimized for the forecasted inflow; (3) the optimized operations are applied for one stage (e.g., one day); and (4) a new meteorological forecast is received and the process is repeated. Yet, the hydrological processes that transform meteorology into river flow are complex and uncertain themselves. The two-step process of optimizing hydrological model parameters to historical observations and then optimizing reservoir operations to model-forecasted river flow may inadvertently yield suboptimal operating policies. This project proposes an alternative, data-driven approach—rather than attempting to model the complex dynamics between weather forecasts/measurements and reservoir river inflow, the data-driven approach will learn these dynamics from data (see Figure 2, pdf attached). That is, a Data-Driven Predictive Controller (DDPC) will use the learned dynamics to make reservoir release decisions. Specifically, the proposed DDPC controller will use a Transient-Predictor DDPC. The Transient Predictor, recently proposed by Institute for Automatic Control (IFA), improves upon prior DDPC methods by making causal multi-step predictions—inputs at timestep k do not affect outputs prior to time k. In addition to the Transient Predictor, proposed DDPC controller will also include an optimal regularization term that has been derived in previous work. This optimal regularization term allows the DDPC to account for system uncertainty due to the noise in the data, favoring control actions that align with the true system dynamics. DDPC reservoir operation is a novel concept and Transient Predictor DDPC is a new control method. Thus, this thesis seeks to make a notable contribution to data-driven reservoir management. A successful thesis may be converted into a publication in a high impact conference or journal. (see pdf for further details)

Goal

Contact Details

More information

Open this project...

Published since: 2024-08-15 , Earliest start: 2024-09-01 , Latest end: 2025-08-31

Applications limited to ETH Zurich

Organization Automatic Control Laboratory

Hosts Wallington Kevin

Topics Engineering and Technology

Optimal Crop Fertilization Control Strategies and Verification

This project deals with the design and analysis of fertilization control strategies. The goal is to minimize over-fertilization while ensuring sufficient nutrification of the crops. Therefore, it is required to study literature on dynamical models of nitrogen in soil, extract a suitable model and implement it in a simulation. Then, design a suitable, formally verifyable control algorithm and analyse the potential of optimal fertilization strategies in agriculture. The control tools may range from dynamic programming (with a-priori guarantees) to reinforcement learning (with statistical a-posteriori guarantees) and beyond.

Keywords

Control Theory, Agriculture, Fertilization, Formal Methods, Safety, Stochastic Systems, Reinforcement Learning

Labels

Master Thesis , ETH Zurich (ETHZ)

Description

Goal

Contact Details

More information

Open this project...

Published since: 2024-08-13

Applications limited to ETH Zurich

Organization Automatic Control Laboratory

Hosts Schmid Niklas

Topics Mathematical Sciences , Information, Computing and Communication Sciences

Data Driven Control Approach for Recommender System Design

The objective of this project is the design and analysis of a smart recommender system as a dynamic feedback controller that, given (some of) the opinions in the system (measured outputs), provides news (namely, the control input) which is tailored to it. The recommender system objective is to optimize his performances, e.g., to maximize engagement, reduce polarization, or robustify against malicious agents. In contrast to other works, we will incorporate learning into this design, using methods from Data-Driven Control.

Keywords

Recommender Systems, Data Driven Control

Labels

Master Thesis , ETH Zurich (ETHZ)

More information

Open this project...

Published since: 2024-07-29 , Earliest start: 2024-03-01

Organization Automatic Control Laboratory

Hosts Eising Jaap , De Pasquale Giulia

Topics Information, Computing and Communication Sciences , Engineering and Technology

Strategically Robust Nash Equilibria in the wild

Game Theory provides the tools to predict and explain the behavior of rational agents that face decision problems when the outcome depends on the decisions of all players. In particular, the concept of Nash Equilibrium is the standard solution concept in this domain. However, there are plenty of examples where people do not behave according to what the theory predicts. In many cases, what the theory predicts (Nash Equilibrium) is clearly not the desired solution, as it is fragile to uncertainty in the game. In the game in the figure, one Nash Equilibrium is that the car maintains its speed hoping that the pedestrian will wait, and another Nash Equilibrium is that the pedestrian crosses hoping that the car stops! On the other hand, agents are also not being robust to the worst case scenario, as that would often lead to no decision being taken at all. In the example in the figure, the security strategy is that the car stops and the pedestrian does not cross, which is clearly unsatisfactory. We recently proposed the concept of Strategically Robust Nash Equilbrium, which interpolates between the concept of Nash Equilibrium (efficient but fragile) and security strategies (robust, but inefficient). In the example in the figure, that corresponds to the car slowing down and the pedestrian waiting to cross -- a sensible outcome. The goal of this project is to validate this concept with real data.

Keywords

Game Theory, Nash equilibrium, behavioral Nash Equilibrium, robust decision making

Labels

Semester Project , Master Thesis

PLEASE LOG IN TO SEE DESCRIPTION

More information

Open this project...

Published since: 2024-07-05 , Earliest start: 2024-09-01 , Latest end: 2025-05-31

Organization Automatic Control Laboratory

Hosts Bolognani Saverio

Topics Mathematical Sciences

Contact

ETH Zurich
Automatic Control Laboratory
Physikstrasse 3
8092 Zurich
Switzerland