Saman Yazdannik

Researcher in Control Theory & Machine Learning for Robotics

Saman Yazdannik

Research

My research interests lie at the intersection of control theory and machine learning, specifically in robotic and multi-agent systems. Classical control provides stability proofs but struggles with autonomous systems operating in uncertain environments. Pure data-driven methods adapt well to changing dynamics but lack formal performance assurance.

I aim to develop hybrid approaches that combine the best of both worlds, achieving provable performance bounds with real-time adaptation to unknown dynamics.

Robotics Control Theory Multi-Agent Systems Model Predictive Control PINNs

Education

K. N. Toosi University of Technology (KNTU)

M.Sc. in Flight Dynamics and Control

  • GPA: 4.0/4.0
  • Thesis: "Behavior-Based Cooperative Control of Multi-Rotors with Collision Avoidance"
  • Advisor: Prof. Morteza Tayefi

Islamic Azad University, Najafabad Branch (IAUN)

B.Sc. in Aerospace Engineering

  • CGPA: 3.25/4.0
  • Thesis: "Control of a Swarm of Aerial Robots with Bifurcation Theory"
  • Advisors: Prof. S.A. Bagherzadeh & Prof. E. KianPour

Experience

Head of Statistics & Graduate Research Assistant

ECCIMA & Univ. of Isfahan Aeronautical Research Center

  • Built data pipelines for departmental performance monitoring and optimized workflows using ML and control methods.
  • Developing optimal control frameworks for space systems under uncertainty.
  • Integrating machine learning with classical control to enhance adaptability and robustness.

Machine Learning Engineer

AFLAK

  • Collaborated on ITU-sponsored project for space applications with Dr. Reza Esmaelzadeh.
  • Fine-tuned large language models for space regulations and compliance.
  • Developed GenAI solutions for space industry applications.

Graduate Research Assistant

KNTU Institute of Intelligent Control Systems

  • Engineered robust payload control for a novel medical delivery aerial robot.
  • Developed cooperative control algorithms for efficient, dynamic task allocation in multi-agent systems.
  • Researched hybrid models to achieve provably safe collision avoidance.

Multi-Agent Autonomy Researcher

Vira Sanaat Kahroba

  • Developed distributed control algorithms for autonomous systems.
  • Implemented multi-agent coordination protocols.

Publications

Awards

  • Best Thesis of the academic year 2023-2024 (2025)
  • M.Sc. with Highest Distinction (Summa Cum Laude) - Top 1% (2021–2024)
  • B.Sc. with High Honors (Magna Cum Laude) - Top 10% (2016–2020)
  • Ranked 104th in National M.Sc. Entrance Examination among 9,000+ candidates (2020)
  • Best Student Magazine Award - Editor-in-Chief, Proxima Engineering Magazine (2020)
  • Ranked 14th in Iran's National Elites Foundation Entrance Examination (2010)