Samer Said Saab
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
Samer Said Saab has a total of 65 co-authors. Here are the top co-authors by number of shared publications:
- Dong Shen
Renmin University of China
8 shared publications - Rayana H. Jaafar
Department of Electrical and Computer Engineering, Lebanese American University, Byblos, Lebanon
5 shared publications - Hao Jiang
Renmin University of China
3 shared publications - Mohammad Al Abbas
Population Studies Center, University of Pennsylvania, Philadelpha, PA, 19104, United States of America
3 shared publications - Zeyi Zhang
School of Mathematics, Renmin University of China, Beijing, China
3 shared publications Khaled Kamal Saab
Stanford University, Palo Alto, California, USA
2 shared publications
Latest Publications
- 16 October 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Industrial Electronics
- Vol. 73 (2) , 3199-3210
Data-Driven Iterative Predictive Control for Unknown Dynamic Systems
This article investigates the challenges associated with precise tracking in industrial batch process control when system dynamics remain unknown. Practical limitations, such as data contamination by noise, insufficient prior knowledge, and sample sparsity, impede the effective deployment of parameterized data-driven methods. To address these challenges, we propose a Gaussian process (GP)-based iterative predictive control (GPIPC) framework. The GPIPC framework leverages GPs to approximate system dynamics from noisy data while explicitly accounting for input noise. It iteratively updates inputs for subsequent batches through predictive control, integrating prediction errors into constraints. A specialized GP update mechanism is developed to facilitate active learning in regions of high prediction variance while reducing interference from outliers. Furthermore, the performance and convergence properties of the proposed scheme are rigorously analyzed via theoretical methods, and we demonstrate its practical feasibility on both simulated and physical robotic platforms.
- 16 October 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Industrial Electronics
- Vol. 73 (2) , 3199-3210
Data-Driven Iterative Predictive Control for Unknown Dynamic Systems
This article investigates the challenges associated with precise tracking in industrial batch process control when system dynamics remain unknown. Practical limitations, such as data contamination by noise, insufficient prior knowledge, and sample sparsity, impede the effective deployment of parameterized data-driven methods. To address these challenges, we propose a Gaussian process (GP)-based iterative predictive control (GPIPC) framework. The GPIPC framework leverages GPs to approximate system dynamics from noisy data while explicitly accounting for input noise. It iteratively updates inputs for subsequent batches through predictive control, integrating prediction errors into constraints. A specialized GP update mechanism is developed to facilitate active learning in regions of high prediction variance while reducing interference from outliers. Furthermore, the performance and convergence properties of the proposed scheme are rigorously analyzed via theoretical methods, and we demonstrate its practical feasibility on both simulated and physical robotic platforms.
- Open Access
- 10 September 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Automatic Control
- Vol. 71 (2) , 1358-1365
Enhancement of Iterative Learning Control With a Novel Error-Adaptive Weighting Mechanism
This article introduces an error-adaptive weighting mechanism (EWM) for iterative learning control (ILC), enhancing convergence by incorporating an iteration-dependent weighting matrix into the learning gain. The EWM’s adaptive weights, responsive to error magnitudes, particularly boost the reduction of large errors. The method’s ability to accelerate convergence while ensuring perfect tracking is rigorously proven. The EWM, being data-driven and easily integrable with existing ILC schemes, is validated through numerical simulations comparing typical ILC schemes and their EWM-enhanced versions.
- 10 September 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Automatic Control
- Vol. 71 (2) , 1358-1365
Open AccessEnhancement of Iterative Learning Control With a Novel Error-Adaptive Weighting Mechanism
This article introduces an error-adaptive weighting mechanism (EWM) for iterative learning control (ILC), enhancing convergence by incorporating an iteration-dependent weighting matrix into the learning gain. The EWM’s adaptive weights, responsive to error magnitudes, particularly boost the reduction of large errors. The method’s ability to accelerate convergence while ensuring perfect tracking is rigorously proven. The EWM, being data-driven and easily integrable with existing ILC schemes, is validated through numerical simulations comparing typical ILC schemes and their EWM-enhanced versions.
- 18 June 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Cybernetics
- Vol. 55 (8) , 3935-3947
Noisy Error-Adaptive Weighting Strategy for Accelerating ILC in Discrete-Time Systems
This article proposes a strategy to accelerate the convergence of iterative learning control (ILC) while maintaining robustness against stochastic noise. The strategy adaptively reweights the error signals used in conventional ILC schemes, casting greater influence to larger errors during input updates, thereby accelerating the correction of noisy inputs and improving overall convergence behavior. Furthermore, to mitigate the impact of noise-dominated small errors on weight computation, a saturation mechanism is introduced. A convergence theorem is established to characterize how the saturation parameters affect the asymptotic convergence of the input deviation-induced errors. Simulation and experimental results demonstrate that incorporating this strategy consistently improves convergence speed while maintaining tracking accuracy across different ILC implementations.
- 18 June 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Cybernetics
- Vol. 55 (8) , 3935-3947
Noisy Error-Adaptive Weighting Strategy for Accelerating ILC in Discrete-Time Systems
This article proposes a strategy to accelerate the convergence of iterative learning control (ILC) while maintaining robustness against stochastic noise. The strategy adaptively reweights the error signals used in conventional ILC schemes, casting greater influence to larger errors during input updates, thereby accelerating the correction of noisy inputs and improving overall convergence behavior. Furthermore, to mitigate the impact of noise-dominated small errors on weight computation, a saturation mechanism is introduced. A convergence theorem is established to characterize how the saturation parameters affect the asymptotic convergence of the input deviation-induced errors. Simulation and experimental results demonstrate that incorporating this strategy consistently improves convergence speed while maintaining tracking accuracy across different ILC implementations.
- 10 November 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE/CAA Journal of Automatica Sinica
- Vol. 11 (1) , 205-218
Data-Driven Learning Control Algorithms for Unachievable Tracking Problems
For unachievable tracking problems, where the system output cannot precisely track a given reference, achieving the best possible approximation for the reference trajectory becomes the objective. This study aims to investigate solutions using the P-type learning control scheme. Initially, we demonstrate the necessity of gradient information for achieving the best approximation. Subsequently, we propose an input-output-driven learning gain design to handle the imprecise gradients of a class of uncertain systems. However, it is discovered that the desired performance may not be attainable when faced with incomplete information. To address this issue, an extended iterative learning control scheme is introduced. In this scheme, the tracking errors are modified through output data sampling, which incorporates low-memory footprints and offers flexibility in learning gain design. The input sequence is shown to converge towards the desired input, resulting in an output that is closest to the given reference in the least square sense. Numerical simulations are provided to validate the theoretical findings.
- 10 November 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE/CAA Journal of Automatica Sinica
- Vol. 11 (1) , 205-218
Data-Driven Learning Control Algorithms for Unachievable Tracking Problems
For unachievable tracking problems, where the system output cannot precisely track a given reference, achieving the best possible approximation for the reference trajectory becomes the objective. This study aims to investigate solutions using the P-type learning control scheme. Initially, we demonstrate the necessity of gradient information for achieving the best approximation. Subsequently, we propose an input-output-driven learning gain design to handle the imprecise gradients of a class of uncertain systems. However, it is discovered that the desired performance may not be attainable when faced with incomplete information. To address this issue, an extended iterative learning control scheme is introduced. In this scheme, the tracking errors are modified through output data sampling, which incorporates low-memory footprints and offers flexibility in learning gain design. The input sequence is shown to converge towards the desired input, resulting in an output that is closest to the given reference in the least square sense. Numerical simulations are provided to validate the theoretical findings.
- 18 July 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Control Systems
- Vol. 43 (4) , 37-39
Samer S. Saab [People in Control]
Q. How did your education and early career lead to your initial and continuing interest in the control field?
- 18 July 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Control Systems
- Vol. 43 (4) , 37-39
Samer S. Saab [People in Control]
Q. How did your education and early career lead to your initial and continuing interest in the control field?
- Open Access
- 30 June 2022
- Published by Wiley in Journal of Robotics
- Vol. 2022, 1-14
Autonomous Robotic Manipulation: Real-Time, Deep-Learning Approach for Grasping of Unknown Objects
Recent advancement in vision-based robotics and deep-learning techniques has enabled the use of intelligent systems in a wider range of applications requiring object manipulation. Finding a robust solution for object grasping and autonomous manipulation became the focus of many engineers and is still one of the most demanding problems in modern robotics. This paper presents a full grasping pipeline proposing a real-time data-driven deep-learning approach for robotic grasping of unknown objects using MATLAB and convolutional neural networks. The proposed approach employs RGB-D image data acquired from an eye-in-hand camera centering the object of interest in the field of view using visual servoing. Our approach aims at reducing propagation errors and eliminating the need for complex hand tracking algorithm, image segmentation, or 3D reconstruction. The proposed approach is able to efficiently generate reliable multi-view object grasps regardless of the geometric complexity and physical properties of the object in question. The proposed system architecture enables simple and effective path generation and a real-time tracking control. In addition, our system is modular, reliable, and accurate in both end effector path generation and control. We experimentally justify the efficacy and effectiveness of our overall system on the Barrett Whole Arm Manipulator.
- 30 June 2022
- Published by Wiley in Journal of Robotics
- Vol. 2022, 1-14
Open AccessAutonomous Robotic Manipulation: Real-Time, Deep-Learning Approach for Grasping of Unknown Objects
Recent advancement in vision-based robotics and deep-learning techniques has enabled the use of intelligent systems in a wider range of applications requiring object manipulation. Finding a robust solution for object grasping and autonomous manipulation became the focus of many engineers and is still one of the most demanding problems in modern robotics. This paper presents a full grasping pipeline proposing a real-time data-driven deep-learning approach for robotic grasping of unknown objects using MATLAB and convolutional neural networks. The proposed approach employs RGB-D image data acquired from an eye-in-hand camera centering the object of interest in the field of view using visual servoing. Our approach aims at reducing propagation errors and eliminating the need for complex hand tracking algorithm, image segmentation, or 3D reconstruction. The proposed approach is able to efficiently generate reliable multi-view object grasps regardless of the geometric complexity and physical properties of the object in question. The proposed system architecture enables simple and effective path generation and a real-time tracking control. In addition, our system is modular, reliable, and accurate in both end effector path generation and control. We experimentally justify the efficacy and effectiveness of our overall system on the Barrett Whole Arm Manipulator.
- Open Access
- 8 March 2022
- Published by EDP Sciences in E3S Web of Conferences
- Vol. 343, 04002
A2L low GWP HFO based refrigerant solutions in heat pump applications to drive energy efficiency, safety, and sustainability even in building restoration
Driven by energy efficiency requirements and legal restrictions like F-Gas Regulation Heat Pump OEM supplier are looking for alternative refrigerant solutions. HFO based low GWP refrigerants can help increasing the system efficiency, expanding the operating temperature range while reducing CO2 emissions. For building restoration in particular, high flow temperatures (for existing hydronic heating system) are essential for convenient and cost-and energy-efficient replacements of fossil fuel boilers. A2L classified refrigerants provide extended application range at highest safety level versus A3 classified refrigerants, but R-32 still carries a Global Warming Potential (GWP) of 675 making it a transitional solution. This paper compares different relevant parameters and points out the advantages provided by HFO A2L low GWP refrigerants (like R-454B and R-454C) for a wide capacity range of air to water and brine to water heat pumps. A special focus is put on the application of building restoration.
- 8 March 2022
- Published by EDP Sciences in E3S Web of Conferences
- Vol. 343, 04002
Open AccessA2L low GWP HFO based refrigerant solutions in heat pump applications to drive energy efficiency, safety, and sustainability even in building restoration
Driven by energy efficiency requirements and legal restrictions like F-Gas Regulation Heat Pump OEM supplier are looking for alternative refrigerant solutions. HFO based low GWP refrigerants can help increasing the system efficiency, expanding the operating temperature range while reducing CO2 emissions. For building restoration in particular, high flow temperatures (for existing hydronic heating system) are essential for convenient and cost-and energy-efficient replacements of fossil fuel boilers. A2L classified refrigerants provide extended application range at highest safety level versus A3 classified refrigerants, but R-32 still carries a Global Warming Potential (GWP) of 675 making it a transitional solution. This paper compares different relevant parameters and points out the advantages provided by HFO A2L low GWP refrigerants (like R-454B and R-454C) for a wide capacity range of air to water and brine to water heat pumps. A special focus is put on the application of building restoration.
- 24 August 2021
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Automatic Control
- Vol. 67 (8) , 4123-4130
Noisy-Output-Based Direct Learning Tracking Control With Markov Nonuniform Trial Lengths Using Adaptive Gains
In this article, a noisy-output-based direct learning tracking control is proposed for stochastic linear systems with nonuniform trial lengths. The iteration-varying trial length is modeled using a Markov chain for demonstration of the iteration dependence. The effect of the noisy output is asymptotically eliminated using a prior given decreasing gain sequence in the learning algorithm. Two alternative adaptive gains are presented for improving the tracking performance and the convergence speed. Both the mean-square and almost-sure convergence are provided. Numerical simulations on a four-degree-of-freedom robot arm are presented to illustrate the effectiveness of the proposed scheme.
- 24 August 2021
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Automatic Control
- Vol. 67 (8) , 4123-4130
Noisy-Output-Based Direct Learning Tracking Control With Markov Nonuniform Trial Lengths Using Adaptive Gains
In this article, a noisy-output-based direct learning tracking control is proposed for stochastic linear systems with nonuniform trial lengths. The iteration-varying trial length is modeled using a Markov chain for demonstration of the iteration dependence. The effect of the noisy output is asymptotically eliminated using a prior given decreasing gain sequence in the learning algorithm. Two alternative adaptive gains are presented for improving the tracking performance and the convergence speed. Both the mean-square and almost-sure convergence are provided. Numerical simulations on a four-degree-of-freedom robot arm are presented to illustrate the effectiveness of the proposed scheme.
- 27 July 2021
- Published by Emerald Publishing in Library Hi Tech
- Vol. 39 (3) , 873-887
Setting the boundaries of COVID-19 lockdown relaxation measures
Purpose: The purpose of this paper is to develop a simple deterministic model that quantifies previously adopted preventive measures driven by the trend of the reported number of deaths in both Italy and India. In addition, the authors forecast the spread based on some selected quantified preventive measures. The optimal exiting policy is derived using the inverse dynamics of the model. Furthermore, the model developed by the authors is dependent on the daily number of deaths; as such, it is sensitive to the death rate but remains insensitive to trends in deaths.Design/methodology/approach: In the wake of COVID-19, policymakers and health professionals realized the limitations and shortcomings of current healthcare systems and pandemic response policies. The need to revise global and national pandemic response mechanisms has been thrust into the public spotlight. To this end, the authors devise an approach to identify the most suitable governmental non-pharmaceutical intervention (NPI) policies, previously adopted in a community, country or region that serve as the foundation for most pandemic strategies.Findings: Leveraging Italy, the authors compare the aftermath by considering three scenarios: (a) recently adopted preventive measures, (b) strictest preventive measures previously adopted, and (c) the optimal exiting policy. In comparison to the second scenario, the authors estimate about twice the number of recoveries and deaths within five months under the first scenario and about 80 times more under the optimal scenario. Whereas in India, the authors applied one scenario of recently adopted preventative measures to showcase the rapid turnaround of their model. According to the new timeline, almost 90% of all deaths in India could have been prevented if the policies implemented in April 2021 were put in place three months prior, i.e. in January 2021.Originality/value: The novelty of the proposed approach is in the use of inverse dynamics of a simple deterministic model that allows capturing the trend of contact rate as a function of adopted NPIs, regardless of pandemic type.
- 27 July 2021
- Published by Emerald Publishing in Library Hi Tech
- Vol. 39 (3) , 873-887
Setting the boundaries of COVID-19 lockdown relaxation measures
Purpose: The purpose of this paper is to develop a simple deterministic model that quantifies previously adopted preventive measures driven by the trend of the reported number of deaths in both Italy and India. In addition, the authors forecast the spread based on some selected quantified preventive measures. The optimal exiting policy is derived using the inverse dynamics of the model. Furthermore, the model developed by the authors is dependent on the daily number of deaths; as such, it is sensitive to the death rate but remains insensitive to trends in deaths.Design/methodology/approach: In the wake of COVID-19, policymakers and health professionals realized the limitations and shortcomings of current healthcare systems and pandemic response policies. The need to revise global and national pandemic response mechanisms has been thrust into the public spotlight. To this end, the authors devise an approach to identify the most suitable governmental non-pharmaceutical intervention (NPI) policies, previously adopted in a community, country or region that serve as the foundation for most pandemic strategies.Findings: Leveraging Italy, the authors compare the aftermath by considering three scenarios: (a) recently adopted preventive measures, (b) strictest preventive measures previously adopted, and (c) the optimal exiting policy. In comparison to the second scenario, the authors estimate about twice the number of recoveries and deaths within five months under the first scenario and about 80 times more under the optimal scenario. Whereas in India, the authors applied one scenario of recently adopted preventative measures to showcase the rapid turnaround of their model. According to the new timeline, almost 90% of all deaths in India could have been prevented if the policies implemented in April 2021 were put in place three months prior, i.e. in January 2021.Originality/value: The novelty of the proposed approach is in the use of inverse dynamics of a simple deterministic model that allows capturing the trend of contact rate as a function of adopted NPIs, regardless of pandemic type.
- 15 June 2021
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Neural Networks and Learning Systems
- Vol. 33 (12) , 7559-7573
A Probabilistically Quantized Learning Control Framework for Networked Linear Systems
In this article, we consider quantized learning control for linear networked systems with additive channel noise. Our objective is to achieve high tracking performance while reducing the communication burden on the communication network. To address this problem, we propose an integrated framework consisting of two modules: a probabilistic quantizer and a learning scheme. The employed probabilistic quantizer is developed by employing a Bernoulli distribution driven by the quantization errors. Three learning control schemes are studied, namely, a constant gain, a decreasing gain sequence satisfying certain conditions, and an optimal gain sequence that is recursively generated based on a performance index. We show that the control with a constant gain can only ensure the input error sequence to converge to a bounded sphere in a mean-square sense, where the radius of this sphere is proportional to the constant gain. On the contrary, we show that the control that employs any of the two proposed gain sequences drives the input error to zero in the mean-square sense. In addition, we show that the convergence rate associated with the constant gain is exponential, whereas the rate associated with the proposed gain sequences is not faster than a specific exponential trend. Illustrative simulations are provided to demonstrate the convergence rate properties and steady-state tracking performance associated with each gain, and their robustness against modeling uncertainties.
- 15 June 2021
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Neural Networks and Learning Systems
- Vol. 33 (12) , 7559-7573
A Probabilistically Quantized Learning Control Framework for Networked Linear Systems
In this article, we consider quantized learning control for linear networked systems with additive channel noise. Our objective is to achieve high tracking performance while reducing the communication burden on the communication network. To address this problem, we propose an integrated framework consisting of two modules: a probabilistic quantizer and a learning scheme. The employed probabilistic quantizer is developed by employing a Bernoulli distribution driven by the quantization errors. Three learning control schemes are studied, namely, a constant gain, a decreasing gain sequence satisfying certain conditions, and an optimal gain sequence that is recursively generated based on a performance index. We show that the control with a constant gain can only ensure the input error sequence to converge to a bounded sphere in a mean-square sense, where the radius of this sphere is proportional to the constant gain. On the contrary, we show that the control that employs any of the two proposed gain sequences drives the input error to zero in the mean-square sense. In addition, we show that the convergence rate associated with the constant gain is exponential, whereas the rate associated with the proposed gain sequences is not faster than a specific exponential trend. Illustrative simulations are provided to demonstrate the convergence rate properties and steady-state tracking performance associated with each gain, and their robustness against modeling uncertainties.