Human-Robot Interaction & Control Interfaces

· 2 min read
project

Project Goal

This project advances human-multi-agent interaction by creating intuitive, responsive control frameworks at the intersection of neuroscience, robotics, and machine learning. We integrate physiological intent decoding with distributed control to build scalable, low-cognitive-load interfaces for supervising complex multi-agent systems. Additionally, we apply adversarial machine learning using probing agents to actively discover and identify leader agents within dynamic multi-agent topologies.

Key Contributions

  • EEG-Based Trust Quantification: Utilized electroencephalography (EEG) signals to decode, quantify, and model real-time human trust levels during human-multi-agent interaction.
  • Brain-Machine Interface (BMI) Control: Formulated novel control methodologies leveraging brain-machine interfaces (BMI) for intuitive, direct human supervision of multi-agent systems.
  • Adversarial Leader Identification: Developed the first learning-based framework for automatic leader identification, using deep reinforcement learning to train a physically interactive probing agent within partially observable multi-agent environments.

Media & Experimental Demos

Experimental Setup Neural Control Interface

Selected Publications

(see Publications for a complete list)

  • Extracting Human Levels of Trust in Human-Swarm Interaction using EEG signals
    Jesus Orozco and Panagiotis Artemiadis
    IEEE Transactions on Human-Machine Systems, vol. 54, no. 2, pp. 182–191, April 2024.
    [PDF]

  • Inferring imagined speech using EEG signals: a new approach using Riemannian manifold features
    Chuong H. Nguyen, George K. Karavas, and Panagiotis Artemiadis
    Journal of Neural Engineering, vol. 15, no. 1, 016002, 2018.
    [PDF] [EEG Dataset]

  • Learning Adversarial Policies for Swarm Leader Identification using a Probing Agent
    Stergios Bachoumas and Panagiotis Artemiadis
    IEEE International Conference on Robotics and Automation (ICRA), Atlanta, GA, 2025.
    [PDF]

Funding & Acknowledgments

This work has been supported by the following grants:

  • DARPA: D14AP00068
  • AFOSR: FA9550-14-1-0149, FA9550-18-1-0221, FA9550-18-1-0464
  • NSF: 2014264
  • Any opinions, findings, and conclusions expressed are those of the authors.