Iman Nematollahi

Iman Nematollahi

Postdoc in Robot Learning

University of Freiburg

Biography

I’m a PostDoc at the University of Freiburg with Prof. Dr. Abhinav Valada, and completed my PhD here under Prof. Dr. Wolfram Burgard. My research focuses on robot manipulation and deep learning to enable robots to develop an intuitive understanding of physics in the form of a world model, learned from past unlabeled interactions. This model allows robots to comprehend the 3D dynamics of their environment, predict plausible future outcomes, and acquire a diverse repertoire of adaptive and generalizable skills.

Progress

Interests
  • Robot Learning
  • World Models
  • Reinforcement Learning
Education
  • Postdoc in Robot Learning, 2025

    University of Freiburg

  • PhD in Robot Learning, 2024

    University of Freiburg

  • MSc in Embedded Systems, 2018

    University of Freiburg

Experience

 
 
 
 
 
PostDoc
Apr 2025 – Present Germany
Advisor: Prof. Abhinav Valada
 
 
 
 
 
Robotics Research Intern
Jan 2024 – Jun 2024 United Kingdom
Advisor: Prof. Ingmar Posner
 
 
 
 
 
Ph.D Candidate in Computer Science
Jan 2019 – Dec 2024 Germany
Advisor: Prof. Wolfram Burgard
 
 
 
 
 
Student Research Assistant
Aug 2017 – Oct 2017 Germany
Implementation of Dynamic Obstacle Avoidance for KUKA OmniRob
 
 
 
 
 
Student Research Assistant
Nov 2016 – Oct 2018 Germany
  • Low power hardware implementation of an Early Seizure Detection Convolutional Neural Network algorithm on FPGA
  • Continuous and simultaneous readout of ADC channels using SPI and USB protocols via FPGA
 
 
 
 
 
Electrical Engineer
Mahya Intelligent Transportation Systems
Sep 2012 – Sep 2014 Iran
As a member of Electrical team, I was trained to design, implement and maintain software and hardware requirements of Tehran traffic control devices.
 
 
 
 
 
Student Research Assistant
Shahid Beheshti University - Microprocessors Lab
Mar 2014 – Jun 2014 Iran
Design and Implementation of a novel Class Attendance System using AVR Microcontrollers and RFID Modules (used in the 22nd Iranian Conference on Electrical Engineering).

Publications

Quickly discover relevant content by filtering publications.
LUMOS: Language-Conditioned Imitation Learning with World Models
ICRA 2025 - Atlanta, USA
LUMOS: Language-Conditioned Imitation Learning with World Models
T3VIP: Transformation-based 3D Video Prediction
IROS 2022 - Kyoto, Japan
T3VIP: Transformation-based 3D Video Prediction
Robot Skill Adaptation via Soft Actor-Critic Gaussian Mixture Models
ICRA 2022 - Philadelphia, USA
Robot Skill Adaptation via Soft Actor-Critic Gaussian Mixture Models
Hardware Implementation of a Performance and Energy-optimized Convolutional Neural Network for Seizure Detection
EMBC 2018 - Honolulu, Hawaii

Contact