Curriculum Vitae

Mohammadhossein Malmir received his Bachelor of Science (B.Sc.) degree in Electrical Engineering (Control Systems) from Amirkabir University of Technology (Tehran Polytechnic), Iran in 2014. Afterwards, he continued his studies at Politecnico di Milano, Italy; and received his Master of Science (M.Sc.) degree in Automation and Control Engineering in 2018. He joined the Chair of Robotics, Artificial Intelligence and Real-time Systems as a research assistant in November 2019. He is currently working in the A-IQ Ready project (successor of the AI4DI project), developing learning algorithms to control industrial manipulators and mobile robots for manufacturing and intralogistics tasks.

 

Research Interests

• Model-based Reinforcement Learning
• Robust Reinforcement Learning
• Data-driven Robust Control
• Approximate Model Predictive Control
• Autonomous Manipulation Learning
• Sim2Real Transfer

Thesis: Feel free to contact me if you are interested in these research areas in order to discuss about possible topics.

 

Publications

J. Josifovski; S. Auddy; M. Malmir; J. Piater; A. Knoll; N. Navarro-Guerrero: Continual Domain Randomization. In IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), 2024. [text], [video]

P. Petropoulakis; L. Gräf; M. Malmir; J. Josifovski; A. Knoll: State Representations as Incentives for Reinforcement Learning Agents: A Sim2Real Analysis on Robotic Grasping. In IEEE International Conference on Systems, Man, and Cybernetics (SMC), 2024. [text], [video]

J. Josifovski; M. Malmir; N. Klarmann; B. L. Žagar; N. Navarro-Guerrero; A. Knoll: Analysis of Randomization Effects on Sim2Real Transfer in Reinforcement Learning for Robotic Manipulation Tasks. In IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), 2022. [text], [video]

N. Klarmann*; M. Malmir*; J. Josifovski*; D. Plorin; M. Wagner; A. Knoll: Optimising Trajectories in Simulations with Deep Reinforcement Learning for Industrial Robots in Automotive Manufacturing. In Artificial Intelligence for Digitising Industry Applications, ISBN: 9788770226646 (pp. 35-45), River Publishers, 2021. [text] , [video]

D. Bojadžić; J. Kunze; D. Osmanković; M. Malmir; A. Knoll: Non-Holonomic RRT & MPC: Path and Trajectory Planning for an Autonomous Cycle Rickshaw. arXiv submission, 2021. [text] , [video]

M. Malmir; J. Josifovski; N. Klarmann; A. Knoll: Robust Sim2Real Transfer by Learning Inverse Dynamics of Simulated Systems. 2nd Workshop on Closing the Reality Gap in Sim2Real Transfer for Robotics, Robotics: Science and Systems (R:SS), 2020. [text] , [video]

J. Josifovski; M. Malmir; N. Klarmann; A. Knoll: Continual Learning on Incremental Simulations for Real-World Robotic Manipulation Tasks. 2nd Workshop on Closing the Reality Gap in Sim2Real Transfer for Robotics, Robotics: Science and Systems (R:SS), 2020. [text] , [video]

M. Malmir: LTV-MPC Control of a Drifting Vehicle for Minimum-time Cornering. M.Sc. Thesis, Politecnico di Milano, 2018 [more...]

M. Malmir; M. Baur; L. Bascetta: A Model Predictive Controller for Minimum Time Cornering. International Conference of Electrical and Electronic Technologies for Automotive, 2018 [text]

M. Malmir; S. Boluki; S. Shiry Ghidary: Offensive Positioning Based on Maximum Weighted Bipartite Matching and Voronoi Diagram. RoboCup 2014: Robot World Cup XVIII, Springer-Verlag, 2015 [text]

 

Teaching

• IN2222 Cognitive Systems - Lectures on Reinforcment Learning -- SS20, SS21, SS22, SS23, SS24
• IN0012, IN2106 Cloud-Based Machine Learning in Robotics - Masterpraktikum -- SS22
• IN0012, IN2106 Simulation-Based Machine Learning in Robotics - Masterpraktikum -- WS22/23, SS23, WS23/24, SS24, WS24/25

 

Student Advising

Ongoing:

• Muhammad Reza Ar Razi, Interdisciplinary Project: Design and Development of a Control and Navigation System with Hierarchical Sim2Real Planning Routines for an Automated Guided Vehicle

Completed:

• SeyedNavid MirnouriLangeroudi, Master's Thesis: A Comparison between Language Models with Transfer Learning on Medical Devices Manuals; Jun. 2023
• Tobias Piltz, Master's Thesis: Safe sim2Real Reinforcement Learning for Robotic Grasping; Apr. 2023; Co-advisor: Josip Josifovski
• Ludwig Gräf, Guided Research: Robotic Grasping - Learning from Visual Environment States; Apr. 2023; Co-advisor: Josip Josifovski
• Panagiotis Petropoulakis, Interdisciplinary Project: Evaluation of Vision-based Robotic Grasping using Deep Reinforcement Learning; Apr. 2023; Co-advisor: Josip Josifovski
• Xiongyu Xie, Master's Thesis: Continual State Representation Learning on Randomized Simulations for Sim2real Transfer in Robotic Manipulation; Mar. 2023; Co-advisor: Josip Josifovski
• Johannes Kiwi, Master's Thesis: Cloning Dynamics of Robotic Manipulation via Explicit Model-based Reinforcement Learning; Jun. 2022; Co-advisor: Josip Josifovski
• Daniel Johannes Hettegger, Master's Thesis: Robotic Grasping with Deep Reinforcement Learning using Suboptimal Teacher Policies; May 2022; Co-advisors: Arne Peters, Josip Josifovski
• Marcel Bruckner, Master's Thesis: Vision-Based Continual Reinforcement Learning for Robotic Manipulation Tasks; Mar. 2022; Co-advisors: Zhenshan Bing, Josip Josifovski
• Shruthi Narayani Venkatesh, Master's Thesis: Simulation-Based Hierarchical Reinforcement Learning for Robotic Manipulation Tasks; Apr. 2021; Co-advisors: Noah Klarmann, Josip Josifovski
• Damir Bojadžić, Master's Thesis: Socially Compliant Path Planning in Dynamic Environments; Feb. 2021
• Gerhard-Mathias Konnerth, Master's Thesis: Exploring Application-oriented Methods to Improve CNN-based Segmentation of SEM Microchip Images; Sept. 2020