Research at Systems and Control
Modeling, inference and control in complex dynamical systems
Current projects
Dynamical systems, physical and others, can be described by a variety of mathematical models. We develop methods for modeling and estimation of dynamical systems based on measurements of input/output data. The models can be used for analysis, in order to better understand the properties of the system, or for control, to automatically regulate a process without human interaction. Contact person: Alexander MedvedevAutomatic control and System identification
Many interesting phenomena around us are complex, dynamical and stochastic in nature, and the available data are inherently uncertain. We develop theory and tools for learning, reasoning and acting based on probabilistic models and measured data, methods that allow humans and machines to better understand the surrounding world. Contact person: Thomas SchönStatistical Machine Learning
The goal of signal processing is to extract information from measured quantities, or signals. This broad concept ranges from simple linear filtering of time series to reduce noise, to nonlinear parameter estimation based on high-dimensional data using statistical models. Estimation theory, optimization, and statistics, play central roles. Contact person: Thomas SchönSignal Processing
Important application areas
The theory and methods of dynamical systems, control, identification and signal processing have much to offer research and clinical practice in modern medicine. We develop methods used in the diagnosis, assessment, and treatment of medical conditions, based on dynamical models of physiological and biological systems. Currently we have projects on Parkinson's disease, breast cancer, diabetes and balance impairment. Contact person: Alexander MedvedevBiomedical systems
Water quality and treatment of water is a growing concern around the world. Demands on quality and increasing loads call for optimized operation of wastewater treatment plants. Applied research in automatic control is an important tool in improving the performance of treatment plants. We are developing control and estimation strategies for wastewater treatment plants that improve pollutant removal, reduce the chemicals consumption and yield energy savings. Contact person: Bengt Carlsson Read more: Modelling and automatic control of wastewater treatment plantsWastewater engineering
Recent publications
More comprehensive list of publications
- Rättsliga hinder mot samverkan?: Juridik och ledningsfrågor vid doktorandsamverkan. SNS förlag, Stockholm, 2025.
- Deep networks for system identification: A survey. In Automatica, volume 171, Elsevier, 2025. (DOI, Fulltext, fulltext:print).
- Robust Estimation of the Covariance Matrix From Data With Outliers. In IEEE Open Journal of Signal Processing, volume 5, pp 1061-1072, Institute of Electrical and Electronics Engineers (IEEE), 2024. (DOI, Fulltext, fulltext:print).
- GNN-IDS: Graph Neural Network based Intrusion Detection System. In Proceedings of the 19th International Conference on Availability, Reliability and Security, ARES ’24, New York, NY, USA, 2024. (DOI, External link).
- Artificial intelligence-enabled electrocardiogram for mortality and cardiovascular risk estimation: a model development and validation study. In The Lancet Digital Health, volume 6, number 11, pp e791-e802, Elsevier, 2024. (DOI, Fulltext, fulltext:print).
- Experimental Characterization of a Robust Localization Method Based on UWB Ranging. In 2024 IEEE International Instrumentation and Measurement Technology Conference, I2MTC 2024, IEEE Instrumentation and Measurement Technology Conference, pp 1-5, Institute of Electrical and Electronics Engineers (IEEE), 2024. (DOI).
- Photo-Realistic Image Restoration in the Wild with Controlled Vision-Language Models. In 2024 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops CVPRW 2024: Proceedings, IEEE Computer Society Conference on Computer Vision and Pattern Recognition workshops : [proceedings], pp 6641-6651, Institute of Electrical and Electronics Engineers (IEEE), 2024. (DOI, Preprint at arXiv).
- Controlling Vision-Language Models for Multi-Task Image Restoration. In , The International Conference on Learning Representations (ICLR), Vienna, Austria, 2024. (Preprint at ArXiv, fulltext:print).
- Hallucination Detection in LLMs: Fast and Memory-Efficient Finetuned Models. Northern Lights Deep Learning 2025 (Wallenberg AI, Autonomous Systems and Software Program (WASP)) (Kjell och Märta Beijers Stiftelse) (National Academic Infrastructure for Supercomputing in Sweden [2022-06725]) (EU, Europeiska forskningsrådet [101054643]), 2024. (fulltext).
- Centrality-based Security Allocation in Networked Control Systems. In Lecture Notes in Computer Science, Springer Publishing Company, 2024.