Uni-Logo
You are here: Home Staff Henrich Kolkhorst
Articleactions

Henrich Kolkhorst

Albert-Ludwigs-Universität Freiburg
Technische Fakultät
Autonome Intelligente Systeme
Georges-Köhler-Allee 080
79110 Freiburg i. Br., Germany
Office: Building 080, room 00.023

kolkhorst@informatik.uni-freiburg.de
Phone: +49 761 203-8024

personal website

Research Interests

  • Brain-Machine Interfaces
  • Human-Robot Interaction
  • Brain State Decoding

I am interested in decoding users' brain states in human-robot interaction scenarios and utilize this information to adapt and improve robots. The user should be able to influence both what the robot does and how these actions are performed. The use of brain signals from electroencephalography (EEG) introduces a feedback modality that is non-intrusive and can also be used by impaired users who have problems communicating through other channels.

In order to control the robot, our work on screen-free object selection using brain signals introduces a novel intuitive way to communicate user intent using in-the-scene highlighting of candidates. While this online setup avoids the need for screen-based interfaces as an indirection step, the object-specific differences in signal distributions pose an interesting challenge from a machine learning perspective.

In pursuance of human-compliant robots, the user's judgment of the environment as well as the robot's actions is critical for improving robotic policies. In the context of autonomous driving, our work on decoding the perceived hazardousness in traffic scenarios from EEG offers a step towards augmenting traditional scene perception with high-level human feedback. Considering robotic assistants in manipulation tasks, EEG-based feedback can be utilized to infer trajectory preferences of the user.

Publications

  • A Robust Screen-Free Brain-Computer Interface for Robotic Object Selection
    Henrich Kolkhorst*, Joseline Veit*, Wolfram Burgard and Michael Tangermann
    Frontiers in Robotics and AI, 7, 2020, Research Topic: Advances in the Integration of Brain–Machine Interfaces and Robotic Devices, In Press
    DOI Manuscript Video Supplementary Material BibTeX
  • Learning User Preferences for Trajectories from Brain Signals
    Henrich Kolkhorst, Wolfram Burgard and Michael Tangermann
    International Symposium on Robotics Research (ISRR), Hanoi, Vietnam, October 2019, In Press
    Manuscript arXiv preprint BibTeX
  • Heterogeneity of Event-Related Potentials in a Screen-Free Brain-Computer Interface
    Henrich Kolkhorst, Joseline Veit, Wolfram Burgard and Michael Tangermann
    8th Graz Brain-Computer Interface Conference, Graz, Austria, September 2019, pp. 228–233
    DOI Manuscript BibTeX
  • Influence of User Tasks on EEG-Based Classification Performance in a Hazard Detection Paradigm
    Henrich Kolkhorst, Saku Kärkkäinen, Amund Faller Raheim, Wolfram Burgard and Michael Tangermann
    41st Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC), Berlin, Germany, July 2019, pp. 6758–6761
    DOI Manuscript BibTeX
  • Mining Within-Trial Oscillatory Brain Dynamics to Address the Variability of Optimized Spatial Filters
    Andreas Meinel, Henrich Kolkhorst and Michael Tangermann
    IEEE Transactions on Neural Systems and Rehabilitation Engineering, vol. 27, no. 3, pp. 378–388, March 2019
    DOI BibTeX
  • Guess What I Attend: Interface-Free Object Selection Using Brain Signals
    Henrich Kolkhorst, Michael Tangermann and Wolfram Burgard
    2018 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), pp. 7111–7116, Madrid, Spain, October 2018
    DOI Manuscript BibTeX
  • Decoding Hazardous Events in Driving Videos
    Henrich Kolkhorst, Wolfram Burgard and Michael Tangermann
    7th Graz Brain-Computer Interface Conference, pp. 242–247, Graz, Austria, September 2017
    DOI Manuscript BibTeX
  • Decoding Perceived Hazardousness From User's Brain States to Shape Human-Robot Interaction
    Henrich Kolkhorst, Michael Tangermann and Wolfram Burgard
    HRI Pioneers Workshop, ACM/IEEE International Conference on Human-Robot Interaction, pp. 349–350, Vienna, Austria, March 2017
    DOI Manuscript BibTeX

Datasets

Datasets for the screen-free object selection can be found on a separate page.

Teaching

About me

  • 06/2015 – up to now: Working at the Autonomous Intelligent Systems group of Prof. Dr. Wolfram Burgard and the Brain State Decoding Lab of Dr. Michael Tangermann as a doctoral research assistant
  • 08/2013 – 02/2015: Co-founder/CTO of PriceNow, working on optimization of personalized pricing and product recommendations in e-commerce
  • 10/2011 – 09/2013: Master of Science with distinction, Computer Science with minor in Operations Research, Karlsruhe Institute of Technology
  • 10/2008 – 09/2011: Bachelor of Science with distinction, Computer Science with minor in Operations Research, Karlsruhe Institute of Technology
Benutzerspezifische Werkzeuge