Arash Bahari Kordabad

Arash Bahari Kordabad

Arash Bahari Kordabad
Postdoctoral Researcher
Max Planck Institute for Software Systems
Email: arashbk@mpi-sws.org
Phone: (+47) 48406162; (+49) 155 10684470
Office: Room 515, Building G 26,
Paul-Ehrlich-Straße 26, 67663 Kaiserslautern,
Germany
URL: https://arashbaharik.github.io
https://www.mpi-sws.org/people/arashbk
Google Scholar    ResearchGate    LinkedIn    Twitter    CV

News

  • Aug 2024: Our paper entitled ``Robust Model Predictive Control for Aircraft Intent-Aware Collision Avoidance" is now available on Arxiv. [ Link ]
  • Aug 2024: I will be a speaker on Safe RL for Smart Grids Workshop at SmartGridComm 2024 , Oslo.
  • July 2024: Our paper on "Distributionally Robust Control for Chance-Constrained Signal Temporal Logic Specifications" has been accepted at the 63rd IEEE Conference on Decision and Control (CDC) 2024.
  • Jun 2024: I will present our paper "Control Barrier Functions for Stochastic Systems under Signal Temporal Logic Tasks," at the European Control Conference (ECC), 2024, Stockholm, Sweden.

Research Interests

My research interests primarily lie at the intersection of Markov Decision Processes (MDPs), Economic Model Predictive Control (EMPC), and Reinforcement Learning (RL), with applications in energy systems and autonomous vehicles. Another key focus of my work is providing probabilistic guarantees for stochastic systems under temporal logic specifications. Additionally, my research encompasses distributionally robust optimization, second-order RL algorithms, and data-driven MPC.

Education & Research Experience

MPI Logo Max Planck Institute for Software Systems, Kaiserslautern, Germany
Postdoctoral Researcher May 2023 - present

Topic: "Multi-agent awareness and control with temporal logic specifications"

Supervisor: Prof. Sadegh Soudjani

Project: SymAware

NTNU Logo Norwegian University of Science and Technology (NTNU), Trondheim, Norway
Ph.D., Department of Engineering Cybernetics Feb 2020 - April 2023

Thesis Topic: "Theoretical properties of learning-based MPC", link

Supervisor: Prof. Sébastien Gros

Co-Supervisor: Prof. Anastasios Lekkas

Committee: Prof. Ole Morten Aamo, Prof. Lars Grüne, and Prof. Rolf Findeisen

AAU Logo Aalborg University, Aalborg, Denmark
Visiting Ph.D. Researcher, Department of Electronic Systems Nov 2021 - Aug 2022

Research topic: Safe Reinforcement Learning

Host: Prof. Rafal Wisniewski

SUT Logo Sharif University of Technology, Tehran, Iran
M.Sc., Department of Mechanical Engineering Sep 2017 - Sep 2019

Thesis Topic: "Control of bifurcation and chatter suppression in peripheral milling process"

Supervisor: Prof. Hamed Moradi

GPA: 19.41/20

UOT Logo University of Tabriz, Tabriz, Iran
B.Sc., Department of Mechanical Engineering Sep 2013 - Sep 2017

Thesis Topic: "On the muscle models as viscoelastic material and comparison of force-length models for active skeletal muscle"

Supervisor: Prof. Kamal Jahani

GPA: 18.1/20

Selected Publications (Top 5)

Selected Talks

  • "Reinforcement Learning for MPC: Fundamentals and Current Challenges" in an invited session entitled Recent Advances in Automated Learning and Calibration of MPC Policies, IFAC, +75 audience, Japan. (photo) July 2023
  • "Introduction to optimization with temporal logic", NTNU, Trondheim, Norway. Mar 2023
  • "Intersection of Reinforcement Learning and MPC", Eindhoven University of Technology, Netherlands, Host: Prof. Dinesh Krishnamoorthy. Feb 2023
  • "MPC-based Reinforcement learning", École polytechnique fédérale de Lausanne (EPFL), Lausanne, Switzerland, Host: Prof. Alireza Karimi. Jan 2023
  • "Optimality Equivalence of MDP and MPC", KTH, Stockholm, Sweden, Host: Prof. Bo Wahlberg. Jan 2023
  • "Intelligent control for time-delay systems", KTH, Stockholm, Sweden, Host: Prof. Håkan Hjalmarsson. Jun 2019