Related Experiment Video
Updated: Jan 5, 2026
11:01
SSVEP-based Experimental Procedure for Brain-Robot Interaction with Humanoid Robots
Published on: November 24, 2015
13.6K
Hybrid Brain-Computer-Interfacing for Human-Compliant Robots: Inferring Continuous Subjective Ratings With Deep
Lukas D J Fiederer1,2, Martin Völker1,2,3, Robin T Schirrmeister1,2
1Neuromedical AI Lab, Department of Neurosurgery, Medical Center - University of Freiburg, Faculty of Medicine, University of Freiburg, Freiburg, Germany.
Frontiers in Neurorobotics
|October 26, 2019
Summary
This study introduces a novel rating system for evaluating robot behavior during human-robot interaction using a hybrid brain-computer interface (BCI). Continuous user feedback allows for adaptive robot responses, enhancing assistive robotic systems.
Area of Science:
- Robotics
- Human-Robot Interaction
- Neuroscience
- Machine Learning
Background:
- Developing human-compliant assistive robots requires appropriate robot behavior during interaction.
- Evaluating the quality of robotic behavior in real-time is crucial for adaptive systems.
Purpose of the Study:
- To develop a continuous rating system for evaluating robot behavior in a hybrid brain-computer interfacing (BCI) task.
- To use collected data to adapt robot behavior based on subjective user perception.
Main Methods:
- A thumb-based wireless controller rating system compatible with dry electroencephalography (EEG) recordings was developed.
- Continuous ratings were collected alongside EEG, respiration, electrocardiogram (ECG), and robotic joint angles.
- Various regression techniques, including deep convolutional neural networks (CNNs), were employed to predict subjective ratings.
Main Results:
- Feasible to obtain continuous rating data offering insights into subjective user perception during human-robot interaction.
- Robot hand position was a better predictor of subjective ratings than EEG, ECG, or respiration.
- Models learned a combination of general and user-specific features, with pre-trained regressors showing accuracy in experienced users.
Conclusions:
- The developed rating system provides a viable method for assessing subjective user experience in human-robot interaction.
- Robot behavior adaptation based on continuous feedback is achievable and can be enhanced by user experience levels.
- Further studies with more participants are needed to validate the methodology for practical application.
