Skip to main content

Stergios E. Bachoumas

Stergios E. Bachoumas

About​

I am a Ph.D. candidate in the Human Oriented Robotics and Control (HORC) lab at the University of Delaware (UD). I received my Diploma in Naval Architecture and Marine Engineering from the National Technical University of Athens (NTUA) in 2019 and my M.S. in Robotics from the University of Delaware in 2025. My research focuses on enhancing the physical intelligence of autonomous robots. My work enables diverse robotic systems to perform complex tasks, ranging from using an autonomous ground vehicle as a probing agent for swarm leader identification to developing a whole-body loco-manipulation controller for quadruped robots.

Useful Information

My Google Scholar

My LinkedIn

My Github

EDUCATION​

Ph.D. in Mechanical Engineering | Sep 2022 - Present University of Delaware (UD)

  • Thesis: "Probing Physical Interactions: Applications in Robotics."

M.S. in Mechanical Engineering | April 2022 National Technical University of Athens (NTUA)

  • Thesis: "Integration of an EtherCAT enabled IMU for the State Estimation of a Quadruped Robot"

B.S. & M.S. in Naval Architecture and Marine Engineering | March 2019 National Technical University of Athens (NTUA)

  • Thesis: "Hybrid Diesel Electric Marine Propulsion Plant Control with Neural Networks"

EXPERIENCE​

Graduate Researcher | Aug 2022 - Present University of Delaware (UD) - (Advisor: Dr. Panagiotis Artemiadis)

  • Created a ROS2 package to integrate an Optitrack motion capture system for robot localization in the testbed.
  • Developed a GPU enabled simulation environment using JAX in Python to facilitate swarm research at scale.
  • Trained and deployed in the real world Deep Reinforcement learning policies for swarm leader identification by introducing a novel Graph Neural Networks (GNNs) layer.
  • Created a framework to train, evaluate and deploy Deep Reinforcement Learning policies for quadrupedal locomotion using PyTorch and ONXX runtime. Performed real world experiments on the Unitree GO1 robot.

Graduate Researcher | Aug 2021 - June 2022 National Technical University of Athens (NTUA) - (Advisor: Dr. Evangelos Papadopoulos)

  • Designed, manufactured and experimentally validated an IMU sensor as an EtherCAT SubDevice Controller.
  • Implemented a tightly coupled Mahony Filter for the attitude estimation of the base of a quadruped robot.

SKILLS​

  • Methods: Deep Reinforcement Learning, Quadrupedal Locomotion, Bayesian Estimation, Multi-robot teams.
  • Software: C/C++(Embedded), Python (Torch, Jax, TensorFlow), Linux/RTOS, ROS/ROS2, Simulation (IsaacLab, Mujoco/MJX, Genesis), Docker, Git, Ansible.
  • Hardware: MCUs (Texas-Instruments/STM32/ESP32), Optitrack MoCap, EtherCAT, PCB Design, Soldering.

PUBLICATIONS​

  1. S. E. Bachoumas, P. Artemiadis. SOMA: Stack-of-Objects Manipulation using a Quadruped Robot. In IEEE Robotics and Automation Letters (RA-L), (In preparation).
  2. S. E. Bachoumas, John Cahill, Panagiotis Artemiadis. Head-On: Robust Tactile Force Sensing for Quadrupedal Interaction. In 2026 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), 2026 PDF.
  3. S. E. Bachoumas, P. Artemiadis. On Swarm Leader Identification using Probing Policies. In IEEE Transactions on Automation Science and Engineering (T-ASE), (Under review) Video.
  4. Jesus A Orozco, S. E. Bachoumas, P. Artemiadis. Quantifying Human Trust in Controlling Robot Swarms: EEG-based Analysis and Classification. In IEEE Engineering in Medicine and Biology Society (EMBC), 2025.
  5. S. E. Bachoumas, P. Artemiadis. Learning Adversarial Policies for Swarm Leader Identification using a Probing Agent. In IEEE International Conference on Robotics and Automation (ICRA), 2025 Video-Paper-PDF.

PROJECTS​

Graduate Mentoring | Oct 2022 - Present​

University of Delaware (UD)

  • Yashwanth Tekumudi (M.Sc. Robotics): An experimental testbed for multi-robot research using ten (10) ground vehicles controlled using ROS2.
  • Camryn Scully (M.Sc. Robotics): A Deep RL locomotion policy for a quadruped over compliant terrains.
  • John Cahill (B.Sc. Mech Eng): A tactile sensing device for a quadruped robot.

Electronics Team Lead | Oct 2018 - Oct 2019​

NTUA Oceanos Boat Racing Team

  • Led the 5-person electronics team during the Monaco Electric Class Boat challenge race in Monte Carlo.
  • Developed the ROS-based telemetry system for the human piloted vessel we designed for the race.
  • Created a PID controller for the speed control of the vessel's propeller.