Mohammad Khosravi
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Research Subjects
Mohammad Khosravi has a total of 54 co-authors. Here are the top co-authors by number of shared publications:
- Roy S. Smith
Department of Information Technology and Electrical Engineering, ETH Zurich, Zurich, Switzerland
16 shared publications - Andrea Iannelli
University of Stuttgart
5 shared publications - Alisa Rupenyan
ETH Switzerland
4 shared publications - Anilkumar Parsi
Automatic Control Laboratory ETH Zurich Switzerland
4 shared publications - John Lygeros
Automatic Control Laboratory, ETH Zrich, Zrich, Switzerland
4 shared publications - Mingzhou Yin
Automatic Control Laboratory ETH Zurich Switzerland
4 shared publications
Latest Publications
- Open Access
Reduced-observable dynamic mode decomposition for identifying theta and beta activity in working-memory EEG
The analysis of transient oscillatory activity in neural recordings remains challenging for linear methods, particularly when signals are nonlinear, noisy, and distributed across multiple channels. This paper introduces a Reduced-Observable Extended Dynamic Mode Decomposition (RO-eDMD) algorithm using a Random Fourier Features dictionary to characterize high-dimensional data in electroencephalography (EEG) recordings. The method extends the Koopman operator framework by incorporating rank reduction directly in the observable space and employing a randomized feature map that approximates the Gaussian kernel, enabling the capture of transient components without explicit dictionary optimization. To assess the method and demonstrate applicability, a logistic regression model is trained using a working memory paradigm as a test case, using dominant power values extracted via RO-eDMD in the theta band as a predictor and the type of event (encoding vs. fixation) as the dependent variable. We find correlations between oscillatory power and memory encoding events in the analyzed bands. Additionally, we find an increase in correlation associated with theta power and memory load, replicating similar results in the literature. The presented algorithm offers a computationally efficient, interpretable framework applied to a working memory paradigm dataset, but is relevant to the analysis of neural time series from diverse experimental contexts.
Open AccessReduced-observable dynamic mode decomposition for identifying theta and beta activity in working-memory EEG
The analysis of transient oscillatory activity in neural recordings remains challenging for linear methods, particularly when signals are nonlinear, noisy, and distributed across multiple channels. This paper introduces a Reduced-Observable Extended Dynamic Mode Decomposition (RO-eDMD) algorithm using a Random Fourier Features dictionary to characterize high-dimensional data in electroencephalography (EEG) recordings. The method extends the Koopman operator framework by incorporating rank reduction directly in the observable space and employing a randomized feature map that approximates the Gaussian kernel, enabling the capture of transient components without explicit dictionary optimization. To assess the method and demonstrate applicability, a logistic regression model is trained using a working memory paradigm as a test case, using dominant power values extracted via RO-eDMD in the theta band as a predictor and the type of event (encoding vs. fixation) as the dependent variable. We find correlations between oscillatory power and memory encoding events in the analyzed bands. Additionally, we find an increase in correlation associated with theta power and memory load, replicating similar results in the literature. The presented algorithm offers a computationally efficient, interpretable framework applied to a working memory paradigm dataset, but is relevant to the analysis of neural time series from diverse experimental contexts.
- 1 September 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Control Systems Technology
- Vol. 34 (5) , 2721-2732
Scalable Predictive Control for District Heating Networks: A Physics-Guided Koopman Operator Approach
- 1 September 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Control Systems Technology
- Vol. 34 (5) , 2721-2732
Scalable Predictive Control for District Heating Networks: A Physics-Guided Koopman Operator Approach
- Open Access
- 17 August 2026
- Published by Elsevier in Neuroimage: Reports
- Vol. 6 (3) , 100398
Latent persistence in fMRI dynamics during human memory consolidation
- 17 August 2026
- Published by Elsevier in Neuroimage: Reports
- Vol. 6 (3) , 100398
Open AccessLatent persistence in fMRI dynamics during human memory consolidation
- Open Access
- 15 June 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Robotics and Automation Letters
- Vol. 11 (8) , 9819-9826
Efficient Data-Driven Reference Governor Design for Safe Evasive Manoeuvring
This paper presents a novel data-driven Reference Governor with Model Predictive Control, integrating local motion replanning and path following for collision avoidance. Employing a model-free Reference Governor, the proposed framework utilises system knowledge through Bayesian Optimisation to augment predetermined evasive trajectories, minimising pathfollowing errors and simultaneously ensuring obstacle safety margins. A single-track vehicle model in combination with a nonlinear tyre model is used to capture the vehicle's dynamics. The optimised control action is the vehicle steering angle, whilst the Reference Governor optimises parameters of a sigmoid reference signal to minimise the tracking error and guarantee safety with respect to obstacles in emergency manoeuvres. The proposed approach is evaluated on a single lane change using a highfidelity simulation environment, and its performance is compared to a baseline controller integrating path following and obstacle avoidance. The results demonstrate a 14% reduction in safety critical overshoot, maximising obstacle safety distance and a four times lower controller cycle time compared to the baseline. Furthermore, through a robustness analysis, it is demonstrated that the proposed approach is more robust towards model mismatches and perception-based errors, as seen by average 30% and 40% reductions in near-miss and collision rates.
- 15 June 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Robotics and Automation Letters
- Vol. 11 (8) , 9819-9826
Open AccessEfficient Data-Driven Reference Governor Design for Safe Evasive Manoeuvring
This paper presents a novel data-driven Reference Governor with Model Predictive Control, integrating local motion replanning and path following for collision avoidance. Employing a model-free Reference Governor, the proposed framework utilises system knowledge through Bayesian Optimisation to augment predetermined evasive trajectories, minimising pathfollowing errors and simultaneously ensuring obstacle safety margins. A single-track vehicle model in combination with a nonlinear tyre model is used to capture the vehicle's dynamics. The optimised control action is the vehicle steering angle, whilst the Reference Governor optimises parameters of a sigmoid reference signal to minimise the tracking error and guarantee safety with respect to obstacles in emergency manoeuvres. The proposed approach is evaluated on a single lane change using a highfidelity simulation environment, and its performance is compared to a baseline controller integrating path following and obstacle avoidance. The results demonstrate a 14% reduction in safety critical overshoot, maximising obstacle safety distance and a four times lower controller cycle time compared to the baseline. Furthermore, through a robustness analysis, it is demonstrated that the proposed approach is more robust towards model mismatches and perception-based errors, as seen by average 30% and 40% reductions in near-miss and collision rates.
- 6 April 2026
- Published by Elsevier in Control Engineering Practice
- Vol. 173
Model-free control of direct air capture: Optimal rule-based policies tuned via Bayesian optimization
- 6 April 2026
- Published by Elsevier in Control Engineering Practice
- Vol. 173
Model-free control of direct air capture: Optimal rule-based policies tuned via Bayesian optimization
- Open Access
- 12 March 2026
- Published by Elsevier in Control Engineering Practice
- Vol. 172
Data-driven high-performance motion system optimization in practice
- 12 March 2026
- Published by Elsevier in Control Engineering Practice
- Vol. 172
Open AccessData-driven high-performance motion system optimization in practice
- 21 January 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Industry Applications
- Vol. PP (00939994) , 1-10
Control-Oriented Hybrid Modeling of AEM Electrolyzer via Residual Error Correction
Accurate, control-oriented models of Anion Exchange Membrane (AEM) electrolyzers are essential for optimization within energy hubs. A significant challenge is balancing predictive accuracy, physical interpretability, and computational tractability required for real-time control. Standard physics-based models, e.g., RC circuits, while interpretable, fail to capture complex, temperature-dependent dynamics, resulting in significant structured error. This paper proposes a control-oriented hybrid modeling framework using a two-stage residual correction strategy, promoting computational efficiency and robustness with respect to measurement and model uncertainties. First, a double-branch RC circuit is identified via a robust multi-step estimation procedure to serve as an interpretable backbone for the dominant electrochemical dynamics. Second, a shallow Neural Network (NN) is trained to learn only the residual error, thereby correcting for the unmodeled thermal dynamics. Validated on real-world operational data from a commercial electrolyzer, the proposed hybrid model achieves a 92.03% $\rm{fit}$. This represents a 66% absolute improvement over the baseline RC-only model (26.00% $\rm{fit}$) and outperforms both pure black-box deep neural network (DNN) (52.04% $\rm{fit}$) and standard physics-in-loss PINN (74.91% $\rm{fit}$) benchmark methods. The resulting framework provides a high-fidelity, tractable, and interpretable solution suitable for advanced model-based control strategies.
- 21 January 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Industry Applications
- Vol. PP (00939994) , 1-10
Control-Oriented Hybrid Modeling of AEM Electrolyzer via Residual Error Correction
Accurate, control-oriented models of Anion Exchange Membrane (AEM) electrolyzers are essential for optimization within energy hubs. A significant challenge is balancing predictive accuracy, physical interpretability, and computational tractability required for real-time control. Standard physics-based models, e.g., RC circuits, while interpretable, fail to capture complex, temperature-dependent dynamics, resulting in significant structured error. This paper proposes a control-oriented hybrid modeling framework using a two-stage residual correction strategy, promoting computational efficiency and robustness with respect to measurement and model uncertainties. First, a double-branch RC circuit is identified via a robust multi-step estimation procedure to serve as an interpretable backbone for the dominant electrochemical dynamics. Second, a shallow Neural Network (NN) is trained to learn only the residual error, thereby correcting for the unmodeled thermal dynamics. Validated on real-world operational data from a commercial electrolyzer, the proposed hybrid model achieves a 92.03% $\rm{fit}$. This represents a 66% absolute improvement over the baseline RC-only model (26.00% $\rm{fit}$) and outperforms both pure black-box deep neural network (DNN) (52.04% $\rm{fit}$) and standard physics-in-loss PINN (74.91% $\rm{fit}$) benchmark methods. The resulting framework provides a high-fidelity, tractable, and interpretable solution suitable for advanced model-based control strategies.
- 16 June 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Control Systems Letters
- Vol. 9, 1634-1639
Sensor Placement in District Heating Networks Using Frequency-Domain Gramians
District heating networks (DHNs) are essential in providing efficient heating services to urban areas through networked pipes. The performance of these systems critically depends on the strategic placement of thermal storage buffers (actuators) and temperature sensors throughout the network. Due to the inherent slow dynamics of thermal transport, these systems exhibit significant delays and periodic behaviors that necessitate time-varying analysis approaches. This paper presents a frequency-domain framework for optimal actuator and sensor placement in DHNs, focusing on metrics derived from frequential Gramians. We provide rigorous analysis of two key metrics, namely the trace and log-determinant of the frequential Gramian, establishing submodularity properties and performance guarantees for greedy selection algorithms. Our theoretical framework naturally handles both the periodic nature of DHNs and their slow transients, outperforming standard approaches in estimation accuracy.
- 16 June 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Control Systems Letters
- Vol. 9, 1634-1639
Sensor Placement in District Heating Networks Using Frequency-Domain Gramians
District heating networks (DHNs) are essential in providing efficient heating services to urban areas through networked pipes. The performance of these systems critically depends on the strategic placement of thermal storage buffers (actuators) and temperature sensors throughout the network. Due to the inherent slow dynamics of thermal transport, these systems exhibit significant delays and periodic behaviors that necessitate time-varying analysis approaches. This paper presents a frequency-domain framework for optimal actuator and sensor placement in DHNs, focusing on metrics derived from frequential Gramians. We provide rigorous analysis of two key metrics, namely the trace and log-determinant of the frequential Gramian, establishing submodularity properties and performance guarantees for greedy selection algorithms. Our theoretical framework naturally handles both the periodic nature of DHNs and their slow transients, outperforming standard approaches in estimation accuracy.
- 2 May 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Control Systems Technology
- Vol. 33 (5) , 1879-1894
Economic Nonlinear Model Predictive Control of Prosumer District Heating Networks
In this article, we propose an economic nonlinear model predictive control (MPC) algorithm for district heating networks (DHNs). The proposed method features prosumers, multiple producers, and storage systems, which are essential components of 4th-generation DHNs. These networks are characterized by their ability to optimize their operations, aiming to reduce supply temperatures, accommodate distributed heat sources, and leverage the flexibility provided by thermal inertia and storage—each crucial for achieving a fossil-fuel-free energy supply. Developing a smart energy management system to accomplish these goals requires detailed models of highly complex nonlinear systems and computational algorithms able to handle large-scale optimization problems. To address this, we introduce a graph-based optimization-oriented model that efficiently integrates distributed producers, prosumers, storage buffers, and bidirectional pipe flows, such that it can be implemented in a real-time MPC setting. Furthermore, we conducted several numerical experiments to evaluate the performance of the proposed algorithms in closed loop. Our findings demonstrate that the MPC methods achieved up to 9% cost improvement over traditional rule-based controllers while better maintaining system constraints.
- 2 May 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Control Systems Technology
- Vol. 33 (5) , 1879-1894
Economic Nonlinear Model Predictive Control of Prosumer District Heating Networks
In this article, we propose an economic nonlinear model predictive control (MPC) algorithm for district heating networks (DHNs). The proposed method features prosumers, multiple producers, and storage systems, which are essential components of 4th-generation DHNs. These networks are characterized by their ability to optimize their operations, aiming to reduce supply temperatures, accommodate distributed heat sources, and leverage the flexibility provided by thermal inertia and storage—each crucial for achieving a fossil-fuel-free energy supply. Developing a smart energy management system to accomplish these goals requires detailed models of highly complex nonlinear systems and computational algorithms able to handle large-scale optimization problems. To address this, we introduce a graph-based optimization-oriented model that efficiently integrates distributed producers, prosumers, storage buffers, and bidirectional pipe flows, such that it can be implemented in a real-time MPC setting. Furthermore, we conducted several numerical experiments to evaluate the performance of the proposed algorithms in closed loop. Our findings demonstrate that the MPC methods achieved up to 9% cost improvement over traditional rule-based controllers while better maintaining system constraints.
- 4 March 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Control Systems Letters
- Vol. 8, 3488-3493
Off-Policy Temporal Difference Learning for Perturbed Markov Decision Processes
Dynamic Programming suffers from the curse of dimensionality due to large state and action spaces, a challenge further compounded by uncertainties in the environment. To mitigate these issue, we explore an off-policy based Temporal Difference Approximate Dynamic Programming approach that preserves contraction mapping when projecting the problem into a subspace of selected features, accounting for the probability distribution of the perturbed transition probability matrix. We further demonstrate how this Approximate Dynamic Programming approach can be implemented as a particular variant of the Temporal Difference learning algorithm, adapted for handling perturbations. To validate our theoretical findings, we provide a numerical example using a Markov Decision Process corresponding to a resource allocation problem.
- 4 March 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Control Systems Letters
- Vol. 8, 3488-3493
Off-Policy Temporal Difference Learning for Perturbed Markov Decision Processes
Dynamic Programming suffers from the curse of dimensionality due to large state and action spaces, a challenge further compounded by uncertainties in the environment. To mitigate these issue, we explore an off-policy based Temporal Difference Approximate Dynamic Programming approach that preserves contraction mapping when projecting the problem into a subspace of selected features, accounting for the probability distribution of the perturbed transition probability matrix. We further demonstrate how this Approximate Dynamic Programming approach can be implemented as a particular variant of the Temporal Difference learning algorithm, adapted for handling perturbations. To validate our theoretical findings, we provide a numerical example using a Markov Decision Process corresponding to a resource allocation problem.