Current PhD students
- Daniel McCauley: Main supervisor
- Vahid Bagherian: Main supervisor
- Abbas Pasdar: Main supervisor
- Seyyed Reza Jafari: Main supervisor
Current Research Projects
Learning Robot Control from Expert Demonstrations and Self-Experience, ELLIIT, 2026-now
- PI: Farnaz Adib Yaghmaie, Linköping University, Co-PI: Yiannis Karayiannidis, Department of Automatic Control, Lund University
- PhD student: Daniel McCauley (will join in Sep. 2026)
- Project description: Mobile manipulation, i.e., manipulation tasks performed by robots that can move through and interact with large, unconstrained environments, requires learning methods that go beyond fixed workspaces. This project aims to investigate and develop learning-based solutions for mobile manipulation that leverage both expert demonstrations and experiential learning. Learning from expert data offers a fast, safe, and efficient pathway to skill acquisition. In parallel, Reinforcement Learning (RL) facilitates systematic exploration and adaptation for manipulation planning. By integrating the strengths of expert-driven learning and RL-based experience, this research seeks to create robust, data-efficient learning strategies for mobile manipulation that generalize across diverse operational environments and transfer effectively to a wide range of real-world robotic applications.
Foundation models and reinforcement learning: A symphony for general-purpose control, SEDDIT 2025-now
- SEDDIT co-applicant, PI: Farnaz Adib Yaghmaie
- PhD student: Abbas Pasdar
- Project description: One of the main challenges in control is generalization to diverse and unseen tasks. Conventional control methods and modern Reinforcement Learning (RL) approaches have focused on task-specific solutions or a tabula rasa approach. These methods learn to solve one task from scratch without incorporating broad knowledge from other datasets, partly due to the incongruity of data from different systems. As a result, sequential decision-making algorithms struggle with generalization. Additionally, the successful deployment of sequential decision-making algorithms on physical systems requires the development of new learning mechanisms that can make appropriate decisions on the fly, utilizing data collected from other tasks. In this project, we are going to study how decision-making algorithms can 1) adapt to modalities not covered in the datasets, 2) generalize to unseen tasks, and 3) adapt to a specific task. The intersection of foundation models and RL holds tremendous promise for creating powerful control systems that can switch and adapt to a diverse range of tasks.
Online Control in Presence of Adversary, ELLIIT 2022-now
- PI: Farnaz Adib Yaghmaie
- Project description: In many practical applications, the noise in the system is not Gaussian. Indeed, the noise in safety-critical systems might be designed by adversaries to deteriorate the performance of the learning systems. The focus of this project is on developing online control algorithms that can successfully accomplish the task even in the presence of adversarial noises. More specifically, we investigated fully observable linear systems subject to adversarial process noise, where the noise can be stochastic, deterministic, designed to be worst-case, or intended to degrade performance.
Reinforcement Learning for Partially Observable Dynamical Systems with Continuous State and Action Space, Zenith, 2021-now
- PI: Farnaz Adib Yaghmaie
- PhD student:Vahid Bagherian
- Project Description: In today’s fast-paced tech world, there is growing interest in learning algorithms for dynamical systems that can be deployed solely based on sensory data without the need for explicit modeling. Reinforcement learning (RL) is key to handling unknown systems; however, it faces challenges with limited sensor data, often lacking complete state information. The main focus of this project is on designing RL algorithms for partially observable dynamical systems based on input-output data with theoretical guarantees.
Multi-agent Systems, 2013-2019
- Project description: During my Ph.D., my research study focused on distributed control of linear heterogeneous multi-agent systems. More specifically, I obtained necessary and sufficient conditions for a group of linear heterogeneous agents to achieve a desired collective behavior like output regulation, bipartite output regulation, multi-party output regulation and formation control. I also developed RL techniques for distributed control of multi-agent systems.
SLAM and Mobile Robot Navigation, 2009-2011
- Project description: I am also interested in Simultaneous Localization And Mapping (SLAM) in Dynamic Environment Using Grid Based Map. During my master's study, I worked on navigation of a nonholonomic mobile robot in dynamic environment. The primal task of the robot was to do SLAM in a dynamic environment and for this purpose, I proposed algorithms to distinguish between dynamic and static obstacles, and to re-do path planning.
Education and Research Experience
- Docent, Electrical Engineering- Control, May 2025
Linköping University, Sweden - Postdoctoral Researcher, Aug. 2018-Dec 2021
Linköping University, Sweden
Project Title: Reinforcement Learning (RL) for control of dynamical systems
From Aug. 2019 till Aug. 2020, I was on maternity leave. - Postdoctoral Research Fellow, May 2017-Feb. 2018
Singapore University of Technology & Design (SUTD), Singapore
Project Title: Reinforcement Learning (RL) for continuous-time systems - Ph.D., Electrical and Electronic Engineering, Aug. 2013 - Apr. 2017
Nanyang Technological University (NTU), Singapore
Thesis Title: Output Regulation of Linear Heterogeneous Multi-Agent Systems
Recipient of the "best thesis award" - Visiting Ph.D. student, May 2015 - Nov. 2015 and Mar. 2016 - May 2016
Centrale Supélec (now Universite Paris-Saclay), Paris, France - Master's Degree, Electrical Engineering-Control, Sep. 2009 - Sep. 2011
K. N. Toosi University of Technology, Tehran, Iran
Thesis Title: Mobile robot navigation in dynamic environment - Bachelor's Degree, Electrical Engineering-Control, Sep. 2005 - Sep. 2009
K. N. Toosi University of Technology, Tehran, Iran
Thesis Title: Quadruple-tank modeling and simulation in Matlab
My personal homepage is updated more frequently
NEWS
- 2026 Sep: We will give the PhD-level course on Reinforcement Learning, WASP again! Stay tuned.
- 2026 Sep: Daniel McCauley will join our group and will work on the ELLIIT project. Looking forward to your contribution, Daniel!
- 2026 July: Welcome Vahid Bagherian to our group. Looking forward to your contributions, Vahid!
- 2025 Dec: Together with Yiannis Karayiannidis, Lund University, we received an ELLIIT grant for two PhD students: one at LiU and one at Lund. Stay tuned.
- 2025 Sep: I gave a PhD-level course on Advanced Robotics at ISY, Linköping University.
- 2025 Jun: I will chair the linear systems session at ECC, Thessaloniki, Greece!
- 2025 May: I am a Docent!
- 2024 Sep-Dec: I gave a PhD-level course on Reinforcement Learning.
- 2024 Sep-Dec: I was a teacher in the PhD-level course on Reinforcement Learning, WASP.
- 2024 January: Together with other researchers from the Automatic Control and Vehicular Systems divisions at ISY, Linköping University and Uppsala University, we started a competence centre called SEDDIT. Svante Gunnarsson is the director of the centre.
- 2023 Jan-Dec: We published two papers in Transactions on Machine Learning Research (TMLR): See here and here.
- 2022 Sep-Dec: I was a teacher in the PhD-level course on Reinforcement Learning, WASP. The course received an excellent evaluation of 3.9/5 and ranked as the second-best PhD course in WASP.
- 2021 Mar: You can now go through our simple handout about Reinforcement Learning entitled "A Crash Course on RL" on arXiv: It is short, easy to read, and comprehensive!
- 2021 Jan: Check out my GitHub page for a crash course on RL. Find out how to implement RL for problems with continuous and discrete action spaces.
- 2021 Jan: I gave a workshop on RL for control at LiU, Linköping, Sweden, in March.
- 2020 Sep: I received a CENIIT grant!
On
https://doi.org/10.23919/ecc65951.2025.11186952
